From 223cfd4e0a18742b348fab99dec673a4b40c0ef5 Mon Sep 17 00:00:00 2001 From: Davood Saffar Date: Tue, 4 Aug 2026 13:04:15 +0000 Subject: [PATCH] refactor(hipblaslt): convert gfx950 F8B8HS library logic to dict format Convert 12 gfx950 F8B8HS library logic YAML files from list to dict format. # Batch: F8B8HS # Per-file lines (before -> after): gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 11112 -> 8123 gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 19887 -> 14362 gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 19887 -> 14357 gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 31361 -> 22462 gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 11112 -> 8123 gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 19887 -> 14362 gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 19887 -> 14357 gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 31361 -> 22462 gfx950/Origami/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 11112 -> 8074 gfx950/Origami/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 19887 -> 14274 gfx950/Origami/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 19887 -> 14269 gfx950/Origami/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml: 31361 -> 22323 Total: 246741 -> 177548 lines across 12 files. --- ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 19085 +++--- ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 34099 ++++------ ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 34094 ++++------ ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 53673 +++++++--------- ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 19083 +++--- ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 34097 ++++------ ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 34092 ++++------ ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 53671 +++++++-------- ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 19036 +++--- ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 34011 ++++------ ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 34006 ++++------ ...F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml | 53534 +++++++-------- 12 files changed, 176644 insertions(+), 245837 deletions(-) diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index b7ae208beed9..2dc0266a955c 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,11034 +81,8043 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1zEvSrUfniyIPA5Nficsug_rlEaVAGxHNYFMuvmYifqc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1oq7eVBWwzey19A8uIqsDkLEvAC8u4jm_3WdeHmyqNY8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1RipBYLITmgLj7nJiXhyI57h-QdswAlXmvZRQ-t7Yij4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33280 - LdsInitCVgprs: false - LdsNumBytes: 33280 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1iIhyCvIYW4EQ_x86i0IZ86zDKGdESkaJ9YYcm0hSGzI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 57600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 57600 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1eZeOxd9MrHLQR2GNA65EfarGb_PDYi0bDvnnTj3iymw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1fok8NkYtMokxqZZMiJ_iGKUDQvUnmdxY1flvKGmyHyE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI4THos_lFJzWi_Dnch9qAlPGNCCGbpGTJT_pt2NL4ZB8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1pEJy9yEKZBXA9EcPO5rjF-8vFHW5-X8UkNyH2oVn994= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29440 - LdsInitCVgprs: false - LdsNumBytes: 29440 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29440 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1S6e2ViSr3lOZh36puDEQ4ogQb2Kj5RPE7WzL1lq3F0M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1C6jTq6qNnho1ENixE4x2q_23xMvUAX0Q3bKaL_ah5XQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21248 - LdsInitCVgprs: false - LdsNumBytes: 21248 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21248 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1158FxB1URc5cpKzdBy5PKmwBlaNnYvU5pNK2aJTi3FY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29440 - LdsInitCVgprs: false - LdsNumBytes: 29440 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29440 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI163lHbClsVTGPnNSzP8QvKwmggFBp2OvBLf9IoiXlJZAY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16cyaq9MxU7sNbLEazHdXIfoficQy4YOPE4TFqMztilgY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20992 - LdsInitCVgprs: false - LdsNumBytes: 20992 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20992 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16O3R5PeDu0JjUKDdGyCc1CoE20Q-U4WqcAOA_B9UwqIo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1LPyOUj4mpprhT6Es7dZqjCKeRoX1IDrnjJ3fYZl1HPI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1dsiiVyF-YL7QpBTxukjNXBE5_N2gPbhGhfp7ioSoYBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33280 - LdsInitCVgprs: false - LdsNumBytes: 33280 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1w-MN2Mm1J0St8q8s9xFauQAWxA0khy1Pd4WoSoekSlg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1cgKWeOcB-B078ImMnsIo0ZS6zy_3JERNpCWjdkytEyw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16_rKHyj2ip3IiAlJ-oI4Mi3S3Jk5OCWj2H83MWBg-Vss= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20992 - LdsInitCVgprs: false - LdsNumBytes: 20992 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20992 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16Qa_B98mEcapAL2iX2YgOcoJisssJyHDlM9u9CPAkaUM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16EwXWjXKR9ATyJNfH3DyoYi3LQ4H0ugLmKOZi9zzkKhc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29056 - LdsInitCVgprs: false - LdsNumBytes: 29056 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 24832 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 8448 - LdsOffsetMetadata_Blk: 24832 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI15LrCqHq2DyagcJd2X5qTNiDHWQnVHm_AqMqcFDlvmsU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1RS5-E3dcQYltJiaIWglbCOB49DJT8EKGQr29y9ht3jk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1WAM7dQoYboHh0SapFH0XK75zgHRfXzHLhlFGHRFgW8U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1fkykvXwBDcTCeZQcsORsfStQSOevaJqa0Sia79tVqpE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21248 - LdsInitCVgprs: false - LdsNumBytes: 21248 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21248 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16WEwxgg8Y4FSUK3HPuX6yKvRQPHQ0p3bui3ZyCjYgs5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16PYalQdoVElqz6wbixR1EPrpAfE0UNhfwtrN4ZWmHKYs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16XYtEgd7zoKly5GXv2t2pAYLX-e76OQRyOB6Ygozeu0A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24960 - LdsInitCVgprs: false - LdsNumBytes: 24960 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1JWVR-jh-aLg0rrdtghL_Hme5Pul5fnhSrwkCOBuPtIU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 768 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 39680 - LdsInitCVgprs: false - LdsNumBytes: 39680 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 6400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 39680 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1PwJ42uBJtJDXBA5J9eUpFXnSBdG873yhqH9x9xHr9aM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35584 - LdsInitCVgprs: false - LdsNumBytes: 35584 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 2304 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35584 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-ckiACWa5_Hl9q0vruUn7fAZYeT9xW9mnDjkhP2c50s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 768 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 39680 - LdsInitCVgprs: false - LdsNumBytes: 39680 - LdsNumElementsAlignedA: 6400 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 6400 - LdsOffsetB_Blk: 71936 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 39680 - LdsOffsetMetadata_Blk: 71936 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1MnL3tvF_jGDAY_I8anP7ysG0uEcoQof1urIS_BrpX_k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35584 - LdsInitCVgprs: false - LdsNumBytes: 35584 - LdsNumElementsAlignedA: 2304 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2304 - LdsOffsetB_Blk: 67840 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35584 - LdsOffsetMetadata_Blk: 67840 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1arXBxIFzyAilDtP1DqyksxdPITOkluw9jmLU5u0Tt2w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1rXabrsjNwkup_JrxozMVR_3U4GdmTFKff0KArypEYAA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1VT_L439A1JYsr4mdupck3knPO5y9YC1PTbi05SL_Cz0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16zQIQS-CrV5B-hJqIIcfIrW4-99zhrrQJDstr_sFS9t4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16yzgS8-MSN2UCxzxDcSy8jzQVwvB-olaZ82o1J7M6llU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI12hFMSt6lMTI8oWW_7aZTwxjiukY6LB_x-MEbbxd5p3M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI163QE_QBcfCjhjm6YvZW1aBPJMqF6E2nryKoAjsugzB0U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16CqP0rVSFO3WDeB13qT6LUGERdrWprMFCekioRFqiTqQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1h8DSl_KtVz4vyL0LXehy8GQa575LmnT-GS8Sm349l8c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 256 - LVCA: 16 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 70912 - LdsInitCVgprs: false - LdsNumBytes: 70912 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 70912 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1ABd-H42vx6NZOkifL5ShxVhgON-oxRsJ0GyaVxfUhIg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 256 - LSPB: 16 - LVCA: 1 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 70912 - LdsInitCVgprs: false - LdsNumBytes: 70912 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 135424 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 70912 - LdsOffsetMetadata_Blk: 135424 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI160Nuqi97WrEd5kWYgnYYkOYOWZ8I-jc1R0TyAYU3gxQk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 16 - LSPA: 256 - LSPB: 256 - LVCA: 1 - LVCB: 1 - LVPA: 16 - LVPB: 16 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16y2XC2HjmywvRNMRzvdc4xwYo7V4GZ1mS8mdECpGkL28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 16 - LSPA: 128 - LSPB: 256 - LVCA: 2 - LVCB: 1 - LVPA: 8 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI162YLFRDLjA2hwymDeM7kI4wEg1_Ytlga74L2JT4VF0MM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 64 - LSCB: 16 - LSPA: 64 - LSPB: 256 - LVCA: 4 - LVCB: 1 - LVPA: 4 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI1651AQfSOaqL9x7qKXZbPBT7ZWUW0NRJycR149B4OSQTc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 32 - LSPA: 256 - LSPB: 128 - LVCA: 1 - LVCB: 2 - LVPA: 16 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI1674YMvMGkXuXL1Cf1gzLE5KEGPC_Zl12HLRkso0LE7_M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 64 - LSPA: 256 - LSPB: 64 - LVCA: 1 - LVCB: 4 - LVPA: 16 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1vOk7fZIE6d2Crn4j8R3f-gr9gYb6d1i7QRitxgT49e8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 16 - LSPA: 256 - LSPB: 256 - LVCA: 1 - LVCB: 1 - LVPA: 16 - LVPB: 16 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xLaYt4XA53nJMoN4GRDxGCoYkMto7pvV9s9Hmk9M424A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1zEvSrUfniyIPA5Nficsug_rlEaVAGxHNYFMuvmYifqc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1oq7eVBWwzey19A8uIqsDkLEvAC8u4jm_3WdeHmyqNY8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1RipBYLITmgLj7nJiXhyI57h-QdswAlXmvZRQ-t7Yij4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33280 + LdsInitCVgprs: false + LdsNumBytes: 33280 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1iIhyCvIYW4EQ_x86i0IZ86zDKGdESkaJ9YYcm0hSGzI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 57600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 57600 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1eZeOxd9MrHLQR2GNA65EfarGb_PDYi0bDvnnTj3iymw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1fok8NkYtMokxqZZMiJ_iGKUDQvUnmdxY1flvKGmyHyE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI4THos_lFJzWi_Dnch9qAlPGNCCGbpGTJT_pt2NL4ZB8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1pEJy9yEKZBXA9EcPO5rjF-8vFHW5-X8UkNyH2oVn994= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 29440 + LdsInitCVgprs: false + LdsNumBytes: 29440 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29440 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1S6e2ViSr3lOZh36puDEQ4ogQb2Kj5RPE7WzL1lq3F0M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1C6jTq6qNnho1ENixE4x2q_23xMvUAX0Q3bKaL_ah5XQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21248 + LdsInitCVgprs: false + LdsNumBytes: 21248 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21248 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1158FxB1URc5cpKzdBy5PKmwBlaNnYvU5pNK2aJTi3FY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 29440 + LdsInitCVgprs: false + LdsNumBytes: 29440 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29440 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI163lHbClsVTGPnNSzP8QvKwmggFBp2OvBLf9IoiXlJZAY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16cyaq9MxU7sNbLEazHdXIfoficQy4YOPE4TFqMztilgY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 20992 + LdsInitCVgprs: false + LdsNumBytes: 20992 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20992 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16O3R5PeDu0JjUKDdGyCc1CoE20Q-U4WqcAOA_B9UwqIo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1LPyOUj4mpprhT6Es7dZqjCKeRoX1IDrnjJ3fYZl1HPI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1dsiiVyF-YL7QpBTxukjNXBE5_N2gPbhGhfp7ioSoYBo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 33280 + LdsInitCVgprs: false + LdsNumBytes: 33280 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1w-MN2Mm1J0St8q8s9xFauQAWxA0khy1Pd4WoSoekSlg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1cgKWeOcB-B078ImMnsIo0ZS6zy_3JERNpCWjdkytEyw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16_rKHyj2ip3IiAlJ-oI4Mi3S3Jk5OCWj2H83MWBg-Vss= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 20992 + LdsInitCVgprs: false + LdsNumBytes: 20992 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20992 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16Qa_B98mEcapAL2iX2YgOcoJisssJyHDlM9u9CPAkaUM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16EwXWjXKR9ATyJNfH3DyoYi3LQ4H0ugLmKOZi9zzkKhc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29056 + LdsInitCVgprs: false + LdsNumBytes: 29056 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 24832 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 8448 + LdsOffsetMetadata_Blk: 24832 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI15LrCqHq2DyagcJd2X5qTNiDHWQnVHm_AqMqcFDlvmsU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1RS5-E3dcQYltJiaIWglbCOB49DJT8EKGQr29y9ht3jk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1WAM7dQoYboHh0SapFH0XK75zgHRfXzHLhlFGHRFgW8U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1fkykvXwBDcTCeZQcsORsfStQSOevaJqa0Sia79tVqpE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 21248 + LdsInitCVgprs: false + LdsNumBytes: 21248 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21248 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16WEwxgg8Y4FSUK3HPuX6yKvRQPHQ0p3bui3ZyCjYgs5k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16PYalQdoVElqz6wbixR1EPrpAfE0UNhfwtrN4ZWmHKYs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16XYtEgd7zoKly5GXv2t2pAYLX-e76OQRyOB6Ygozeu0A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 24960 + LdsInitCVgprs: false + LdsNumBytes: 24960 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1JWVR-jh-aLg0rrdtghL_Hme5Pul5fnhSrwkCOBuPtIU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 768 + LdsBytesNoAmax: 39680 + LdsInitCVgprs: false + LdsNumBytes: 39680 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 6400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 39680 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1PwJ42uBJtJDXBA5J9eUpFXnSBdG873yhqH9x9xHr9aM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 35584 + LdsInitCVgprs: false + LdsNumBytes: 35584 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 2304 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35584 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-ckiACWa5_Hl9q0vruUn7fAZYeT9xW9mnDjkhP2c50s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 768 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 39680 + LdsInitCVgprs: false + LdsNumBytes: 39680 + LdsNumElementsAlignedA: 6400 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 6400 + LdsOffsetB_Blk: 71936 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 39680 + LdsOffsetMetadata_Blk: 71936 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1MnL3tvF_jGDAY_I8anP7ysG0uEcoQof1urIS_BrpX_k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 35584 + LdsInitCVgprs: false + LdsNumBytes: 35584 + LdsNumElementsAlignedA: 2304 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2304 + LdsOffsetB_Blk: 67840 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35584 + LdsOffsetMetadata_Blk: 67840 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1arXBxIFzyAilDtP1DqyksxdPITOkluw9jmLU5u0Tt2w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1rXabrsjNwkup_JrxozMVR_3U4GdmTFKff0KArypEYAA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1VT_L439A1JYsr4mdupck3knPO5y9YC1PTbi05SL_Cz0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16zQIQS-CrV5B-hJqIIcfIrW4-99zhrrQJDstr_sFS9t4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16yzgS8-MSN2UCxzxDcSy8jzQVwvB-olaZ82o1J7M6llU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI12hFMSt6lMTI8oWW_7aZTwxjiukY6LB_x-MEbbxd5p3M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI163QE_QBcfCjhjm6YvZW1aBPJMqF6E2nryKoAjsugzB0U= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16CqP0rVSFO3WDeB13qT6LUGERdrWprMFCekioRFqiTqQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1h8DSl_KtVz4vyL0LXehy8GQa575LmnT-GS8Sm349l8c= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 256 + LVCA: 16 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 70912 + LdsInitCVgprs: false + LdsNumBytes: 70912 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 70912 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1ABd-H42vx6NZOkifL5ShxVhgON-oxRsJ0GyaVxfUhIg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 256 + LSPB: 16 + LVCA: 1 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 70912 + LdsInitCVgprs: false + LdsNumBytes: 70912 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 135424 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 70912 + LdsOffsetMetadata_Blk: 135424 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI160Nuqi97WrEd5kWYgnYYkOYOWZ8I-jc1R0TyAYU3gxQk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 16 + LSPA: 256 + LSPB: 256 + LVCA: 1 + LVCB: 1 + LVPA: 16 + LVPB: 16 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16y2XC2HjmywvRNMRzvdc4xwYo7V4GZ1mS8mdECpGkL28= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 16 + LSPA: 128 + LSPB: 256 + LVCA: 2 + LVCB: 1 + LVPA: 8 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI162YLFRDLjA2hwymDeM7kI4wEg1_Ytlga74L2JT4VF0MM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 16 + LSPA: 64 + LSPB: 256 + LVCA: 4 + LVCB: 1 + LVPA: 4 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI1651AQfSOaqL9x7qKXZbPBT7ZWUW0NRJycR149B4OSQTc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 32 + LSPA: 256 + LSPB: 128 + LVCA: 1 + LVCB: 2 + LVPA: 16 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI1674YMvMGkXuXL1Cf1gzLE5KEGPC_Zl12HLRkso0LE7_M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 64 + LSPA: 256 + LSPB: 64 + LVCA: 1 + LVCB: 4 + LVPA: 16 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1vOk7fZIE6d2Crn4j8R3f-gr9gYb6d1i7QRitxgT49e8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 16 + LSPA: 256 + LSPB: 256 + LVCA: 1 + LVCB: 1 + LVPA: 16 + LVPB: 16 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xLaYt4XA53nJMoN4GRDxGCoYkMto7pvV9s9Hmk9M424A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index 1c03eb5e754a..76e2c01eab12 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,19809 +81,14282 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIk_RqFM_AwoflFBuOWbvk-8fTqjDZ1FddXgJjbcy-G0k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 125952 - LdsInitCVgprs: false - LdsNumBytes: 125952 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 6] - MIWaveTileA: 8 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 192 - MacroTileA: 256 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 6 - ThreadTileA: 32 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MISPq2JndWaX4OL12RUN6ulsn7QcNRohtTIZC1Edgdyws= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123904 - LdsInitCVgprs: false - LdsNumBytes: 123904 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 5] - MIWaveTileA: 8 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 160 - MacroTileA: 256 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 5 - ThreadTileA: 32 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIKVXPdd7d_qWsOE3R4Nn3WRva2YN_nzMUEkSXMgI4hvI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50176 - LdsInitCVgprs: false - LdsNumBytes: 50176 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50176 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1T7zX-LAq1W4S0p0x8V1ahqv2mXqZFvICIzJD4wjn5vU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48128 - LdsInitCVgprs: false - LdsNumBytes: 48128 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48128 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1C8Q9_g8YRB2HVCaci6ZpxWtz2F-KOjbPdcOSgM2OeEI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1wcQqKyv34c9LbqUML817LBELS8jNDkpatWFuPGnAz0I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37888 - LdsInitCVgprs: false - LdsNumBytes: 37888 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37888 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIghxlKUHytHxV_rsO7Y3k-hKsNgsUv14Ji4M_f4zzLUg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50432 - LdsInitCVgprs: false - LdsNumBytes: 50432 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50432 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 5] - MIWaveTileA: 6 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 160 - MacroTileA: 192 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 5 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 5 - ThreadTileA: 24 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MIJBI5cmbVufum9IvszCzB765s6mkvxNBCOv5r57HrQ5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI195MLYeeCjfNjgEj-ik95KMPTRnsuVyYIPGD8iI7U3FA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1enbINkflafBFO3_Pebg7jg-XrgUNrjPKEtJ1FG_ArW4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1AXbZC25QBPC0PuokhYA5b8WyOR4HvuGr9lZsTR0HMWk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 57600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 57600 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIYn1IItDWJZY2hmD9-0eW5TADd3akO6yu4iNVSScCFLk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 86272 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 86272 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI15qZVVldYFEVxv6DoRCJq1o-8jbrdV19RqRcT-bcTp5g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 86272 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 86272 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1lHcr20MZlAXhwNOR6Gevj4wJaRc4B-tum0bYx49prt8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1XvrvtOqqq8hKEVOCgfEFmALGzVphawbwpkAxG7nqoi8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIYXJbKl66tz5Yz16uScZ-3QcnIi8M-PaVMoZpl5IQJR0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 8] - MIWaveTileA: 4 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 8 - ThreadTileA: 16 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIWMlYxh0gBBwfi1wvTDHJHkSEVXIuF12-m4Tq-srJGFk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 224 - MacroTileA: 128 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIU-MW4z63wAohwJUDz0C8OcgBozd0DUJEBLSSdararIA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIZLtCnUAcZQRb2Sm7BFv7A3a2vBOCNqxyJPub1dc8Z8U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MInMSU2Gh3mMVemcGZ3WqUWu2l4NG8N5rOCZZl_tZSxr4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1VCOEzWL6LPvmD0HkYK2IQmKeQ33MFQ6q3DD1gUWXPWI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32256 - LdsInitCVgprs: false - LdsNumBytes: 32256 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32256 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1bvrvP40X4hFkSo1KwPg0VEiYDXT6P8LD2PVpHtPr4t8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1KCNveOe_jYnymu00u65Iq4hQQmFEfEECKI4hYCSw86A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22016 - LdsInitCVgprs: false - LdsNumBytes: 22016 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22016 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI11tDVdUBSEo3UIS5TLyJqHVgQw3dto3agLBzaHthkf5A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46336 - LdsInitCVgprs: false - LdsNumBytes: 46336 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46336 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1uDCZdlP32w2aUIZQgU6UFaItvoNYlwEGmbTkXc1IEtc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48384 - LdsInitCVgprs: false - LdsNumBytes: 48384 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48384 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 7] - MIWaveTileA: 3 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 224 - MacroTileA: 96 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 3 - NumLoadsB: 7 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 7 - ThreadTileA: 12 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1tIdq_8HHc1AUHlfryN0_49Q8EAD9oEnK2lMgmhIKC2E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1o_9BqNra6zh_HmLvWFNQm4WYzN2W1bygUabZEe9Lzsg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1KePfYaceZ29tiGigvOtcCFJNsP7pueMsjzyHKVb7w64= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16tOU8-XLYrNPdvvuRaPZblzPTCq0IYpmOqRVYwuw2zIA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 27904 - LdsInitCVgprs: false - LdsNumBytes: 27904 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27904 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16iQaL89dfDqFaUwHSFuqBVL1d7gxxRqiMEISWe48rBhk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16G697WmrCkkuSW0ygicOEsjSabf2uABvGDaKBNngK4oE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1opMCZdZFzR31a1nXMExXx-HFGwt6FHBNysNBuXPo34o= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1GPhaxX2PStJ8QfX3S-AgQFNJD6o4d_c2qqsOdt_oaC4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44288 - LdsInitCVgprs: false - LdsNumBytes: 44288 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44288 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1gqbADNhoP3bHba8oMSfqNYB-76ixvzWU8E58C_oXxBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1-VJoMMTo08ubflQERsCoO0j_TzMOoFOBMG1VV9dqzpY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI15OARqHSqH8kbGBPru-B8g6Oad-Zo3FGsUS8SxkzZZYk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16xVRyCL8pcYWjhmBX3Sksf7_ECJZKzPUO0YM7WCFd9sM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 23808 - LdsInitCVgprs: false - LdsNumBytes: 23808 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23808 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI162emMCIrLY9uAeJtwZl6HlANVDofIksu3xpgoJdG4L7I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16UNManzuffvMTmsgfWnWF6Bt5jjvtS0tBsdToDQ_bZvc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 24832 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 8448 - LdsOffsetMetadata_Blk: 24832 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1kHn9o1wGnb0Le03rKuL4bApCYLkFlSLtoU6MjkRibzc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1NMWe8o4PG1BIiKP1ggEjZUpwkkJSDyvBBNMZ8tXiaR8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 1 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1Q0rm-ZJ-B1Ep3tDRV7Y6v5xFppIw1pc1wB6W7qCB6eE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32000 - LdsInitCVgprs: false - LdsNumBytes: 32000 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32000 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI14_XY01zUqMSgVV3Z29zc0HvqY2RBU1XaMixpmKPTFvQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1BV068w-wQamiyACMJrCQO6AzvkFtVVaL6YUOIybq_As= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI167ogjNu2arPxWNgqgKM8sDxj_QbeHe5G3biyDU4kITPE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 19712 - LdsInitCVgprs: false - LdsNumBytes: 19712 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 19712 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI1606uPSrg8jSF_adB8-Q1EUFxmHwf5F9lLMz64s3UmyCk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16Jm6coo6HlXoYc23v3SkS10L-uM9lwa5bnHnY2Vs_oLo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIv9D6qCD557UuGj0WYoB1KLme1Sh1mzoR8zvP-Ew1rDY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 66560 - LdsInitCVgprs: false - LdsNumBytes: 66560 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 66560 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 13] - MIWaveTileA: 4 - MIWaveTileB: 13 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 208 - MacroTileA: 256 - MacroTileB: 208 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 52 - NumLoadsA: 8 - NumLoadsB: 13 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 13 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 13 - ThreadTileA: 16 - ThreadTileB: 13 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIZTzgVMYbF8zAjkLbWPx_KguNjRPGDWzQphgfwu106mI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 28160 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 11] - MIWaveTileA: 4 - MIWaveTileB: 11 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 176 - MacroTileA: 256 - MacroTileB: 176 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 44 - NumLoadsA: 8 - NumLoadsB: 11 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 11 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 11 - ThreadTileA: 16 - ThreadTileB: 11 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIYpXgEPF2KmPhp_HSItkaLdu8ZfTql6wuzRMctpM8bWs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 121856 - LdsInitCVgprs: false - LdsNumBytes: 121856 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 9] - MIWaveTileA: 4 - MIWaveTileB: 9 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 144 - MacroTileA: 256 - MacroTileB: 144 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 8 - NumLoadsB: 9 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 9 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 9 - ThreadTileA: 16 - ThreadTileB: 9 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIM6ky9Tq6VRezat1v-Vy4Qy4OEEUGyFJHwE-XvlIpRaY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 17920 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 112 - MacroTileA: 256 - MacroTileB: 112 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1ZaT28UTzCJBM6wCUWArJk9cu4afCdA6oGhULOQC7jwQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46080 - LdsInitCVgprs: false - LdsNumBytes: 46080 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 12800 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46080 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1ckLt2U3Q4BSlwSpNR-cpd6uJ3sMr5OVPRWuKvEiSVtY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 7680 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1Cks6yoWu7mRHY5OU_L0wlyt_BkdBZ9B6KYLX0jMjEZM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35840 - LdsInitCVgprs: false - LdsNumBytes: 35840 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 2560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1LRieWm8obTv5BrugK6ZfaH9J4-zT3LrRBcx6GxwPdUk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 1280 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45312 - LdsInitCVgprs: false - LdsNumBytes: 45312 - LdsNumElementsAlignedA: 10496 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 10496 - LdsOffsetB_Blk: 76032 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45312 - LdsOffsetMetadata_Blk: 76032 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1zBVp9O_j-pDQmFJ0pruVZjPjyXjbaxIDNKuOKzE-hUk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 768 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 6400 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 6400 - LdsOffsetB_Blk: 71936 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 71936 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1LmEbjkFjRe2EHvxZ5njVJSBpnw78dqd37eC05zoq8oY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 2304 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2304 - LdsOffsetB_Blk: 67840 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 67840 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI_Cpsqq7IdCcI6m09ixnRrSz_HTRJbwY41LkpuZ6eaow= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1tP0yLw5sOaqSqLw69T0rUISjIec4gtRcEBnLindXxd0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1WlKSkQYWc45q9-9u3Sa8KR5G5YSwdMXXvYIooja5mzU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1ZFz2Wqi9AGPfO9YZeT8hu0U-a6QRpBHZ1MNZhOuWfzs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16YTIDOkvWpHTvJMbXMK3Qn2TUcHpLDIz3L67zYqwLtxM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1Hyiuad9E_LwogHYOQgHONBkR83eXTUE_-HwHw1z6c2M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 62976 - LdsInitCVgprs: false - LdsNumBytes: 62976 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 62976 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1_vkhGjNt0PNt_xV8VVOz2XzUCfwEcU8zljVlbKTICVg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16riuLixhIaRRQRAFH27-FM-5iTnYrSUKGh2HFdVNH4W8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16tRLsgniRf9NQ8cwUknJpPeoosk2I0euxrG_yXuRD6Zc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16zCjLhPMzebcvjlA5OUrujTBGN1OYulB3ktj1a570vLs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1OuevmgTEfRQM80_r7hzorlLPNIZiSHCEk0LhjgcOAH8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 60672 - LdsInitCVgprs: false - LdsNumBytes: 60672 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 60672 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI117HfbCeMXp_75IhCNgyVsAtudVa47Y080duX4tlb_Cc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 54528 - LdsInitCVgprs: false - LdsNumBytes: 54528 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 54528 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1w_TVfNrjE3mYvq96Bm1Jgkl-HCbXvzNQ6H6gr4PSnKc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16IX89BCLJtYzlNb-bQG51jUiesgW7gn9abBGKzaC4u7Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16yEC0PHvmD0zrKvRDTkhqsmiXq78VvD0Eg5fcAnocoic= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16_OtB8fTrZk3Pt0bE3bXfsWfkoWXEPa2ENPfe4eaiVRI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1uaMQi1863lUOUemwTE99mttXxpTRkvqvmiwvtfx6Qx0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 80384 - LdsInitCVgprs: false - LdsNumBytes: 80384 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 13824 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 80384 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 16 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1_aftRJpk6u1tTs3-Hwls-Fo3A3dD889Os-hWuvfNQvw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71168 - LdsInitCVgprs: false - LdsNumBytes: 71168 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 4608 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71168 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI18CjcSziK0PjBic2MAopWR-Ta_nHbDse6qNtds5pye7Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 256 - LSPB: 16 - LVCA: 1 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71936 - LdsInitCVgprs: false - LdsNumBytes: 71936 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 135424 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71936 - LdsOffsetMetadata_Blk: 135424 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 512 - LSPA: 64 - LSPB: 8 - LVCA: 4 - LVCB: 32 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI169OaDNKWZnXMa1R_AlPEgQpVqAFVb54TBTn3PvUMxY7A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 512 - LSPA: 128 - LSPB: 8 - LVCA: 2 - LVCB: 32 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16JRL6UviIw3ql2MWJlkjDhAPVUQuM-dfo_YP9V8INeCE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16a9rQBxUkyvDWlmJquyIGWqzMINaNpRD_aWL9QWIuzQ8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 512 - LSPA: 128 - LSPB: 8 - LVCA: 2 - LVCB: 32 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16Jyq1ax9kWbpfAk08jNUVf7z1XoXRVCVdpt-kBpa0fWY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 64 - LSCB: 512 - LSPA: 64 - LSPB: 8 - LVCA: 4 - LVCB: 32 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16D0scrs9_FOYkVHUHqfaWIBeu8Rz72d3pOV3R8iV8H74= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI169Ao_eCjaajChoPx9i0y3MGqMyeJdbqU3dbjwB_UIKx4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1DDaFTcewwV5Ybn8swfgQtfH13aPb98dLOnRrTIJYmSQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1ifWE51xli667eSWHdf8X0Gqm1isofKMCZj08QiVl3U0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 1024 - LSPA: 128 - LSPB: 4 - LVCA: 2 - LVCB: 64 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1U06udU6pkFuvB0exyiBHs-wOgMdXOgbcv4nENk4uw44= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1oDw0RAqdNEFhP7idqV0O87totk9ugWGoiXKFb7JRQVw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84992 - LdsInitCVgprs: false - LdsNumBytes: 84992 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84992 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xqwn4mYe9PqjtRIapRznDsflXKMnaPO3v98RcshyCRoc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 16 - LdsPadB: 8 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIk_RqFM_AwoflFBuOWbvk-8fTqjDZ1FddXgJjbcy-G0k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 125952 + LdsInitCVgprs: false + LdsNumBytes: 125952 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 6] + MIWaveTileA: 8 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 192 + MacroTileA: 256 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 6 + ThreadTileA: 32 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MISPq2JndWaX4OL12RUN6ulsn7QcNRohtTIZC1Edgdyws= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 123904 + LdsInitCVgprs: false + LdsNumBytes: 123904 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 5] + MIWaveTileA: 8 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 160 + MacroTileA: 256 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 5 + ThreadTileA: 32 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIKVXPdd7d_qWsOE3R4Nn3WRva2YN_nzMUEkSXMgI4hvI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 50176 + LdsInitCVgprs: false + LdsNumBytes: 50176 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50176 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1T7zX-LAq1W4S0p0x8V1ahqv2mXqZFvICIzJD4wjn5vU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 48128 + LdsInitCVgprs: false + LdsNumBytes: 48128 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48128 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1C8Q9_g8YRB2HVCaci6ZpxWtz2F-KOjbPdcOSgM2OeEI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1wcQqKyv34c9LbqUML817LBELS8jNDkpatWFuPGnAz0I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 37888 + LdsInitCVgprs: false + LdsNumBytes: 37888 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37888 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIghxlKUHytHxV_rsO7Y3k-hKsNgsUv14Ji4M_f4zzLUg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 50432 + LdsInitCVgprs: false + LdsNumBytes: 50432 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50432 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 5] + MIWaveTileA: 6 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 160 + MacroTileA: 192 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 5 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 5 + ThreadTileA: 24 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MIJBI5cmbVufum9IvszCzB765s6mkvxNBCOv5r57HrQ5k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI195MLYeeCjfNjgEj-ik95KMPTRnsuVyYIPGD8iI7U3FA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1enbINkflafBFO3_Pebg7jg-XrgUNrjPKEtJ1FG_ArW4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1AXbZC25QBPC0PuokhYA5b8WyOR4HvuGr9lZsTR0HMWk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 57600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 57600 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIYn1IItDWJZY2hmD9-0eW5TADd3akO6yu4iNVSScCFLk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 86272 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 86272 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI15qZVVldYFEVxv6DoRCJq1o-8jbrdV19RqRcT-bcTp5g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 86272 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 86272 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1lHcr20MZlAXhwNOR6Gevj4wJaRc4B-tum0bYx49prt8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1XvrvtOqqq8hKEVOCgfEFmALGzVphawbwpkAxG7nqoi8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIYXJbKl66tz5Yz16uScZ-3QcnIi8M-PaVMoZpl5IQJR0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 8] + MIWaveTileA: 4 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 8 + ThreadTileA: 16 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIWMlYxh0gBBwfi1wvTDHJHkSEVXIuF12-m4Tq-srJGFk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 224 + MacroTileA: 128 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 7 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIU-MW4z63wAohwJUDz0C8OcgBozd0DUJEBLSSdararIA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIZLtCnUAcZQRb2Sm7BFv7A3a2vBOCNqxyJPub1dc8Z8U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MInMSU2Gh3mMVemcGZ3WqUWu2l4NG8N5rOCZZl_tZSxr4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1VCOEzWL6LPvmD0HkYK2IQmKeQ33MFQ6q3DD1gUWXPWI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32256 + LdsInitCVgprs: false + LdsNumBytes: 32256 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32256 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1bvrvP40X4hFkSo1KwPg0VEiYDXT6P8LD2PVpHtPr4t8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1KCNveOe_jYnymu00u65Iq4hQQmFEfEECKI4hYCSw86A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 22016 + LdsInitCVgprs: false + LdsNumBytes: 22016 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22016 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI11tDVdUBSEo3UIS5TLyJqHVgQw3dto3agLBzaHthkf5A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 46336 + LdsInitCVgprs: false + LdsNumBytes: 46336 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46336 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1uDCZdlP32w2aUIZQgU6UFaItvoNYlwEGmbTkXc1IEtc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 48384 + LdsInitCVgprs: false + LdsNumBytes: 48384 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48384 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 7] + MIWaveTileA: 3 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 224 + MacroTileA: 96 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 3 + NumLoadsB: 7 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 7 + ThreadTileA: 12 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1tIdq_8HHc1AUHlfryN0_49Q8EAD9oEnK2lMgmhIKC2E= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1o_9BqNra6zh_HmLvWFNQm4WYzN2W1bygUabZEe9Lzsg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1KePfYaceZ29tiGigvOtcCFJNsP7pueMsjzyHKVb7w64= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16tOU8-XLYrNPdvvuRaPZblzPTCq0IYpmOqRVYwuw2zIA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 27904 + LdsInitCVgprs: false + LdsNumBytes: 27904 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27904 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16iQaL89dfDqFaUwHSFuqBVL1d7gxxRqiMEISWe48rBhk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16G697WmrCkkuSW0ygicOEsjSabf2uABvGDaKBNngK4oE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1opMCZdZFzR31a1nXMExXx-HFGwt6FHBNysNBuXPo34o= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1GPhaxX2PStJ8QfX3S-AgQFNJD6o4d_c2qqsOdt_oaC4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 44288 + LdsInitCVgprs: false + LdsNumBytes: 44288 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44288 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 7 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1gqbADNhoP3bHba8oMSfqNYB-76ixvzWU8E58C_oXxBo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1-VJoMMTo08ubflQERsCoO0j_TzMOoFOBMG1VV9dqzpY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI15OARqHSqH8kbGBPru-B8g6Oad-Zo3FGsUS8SxkzZZYk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16xVRyCL8pcYWjhmBX3Sksf7_ECJZKzPUO0YM7WCFd9sM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 23808 + LdsInitCVgprs: false + LdsNumBytes: 23808 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23808 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI162emMCIrLY9uAeJtwZl6HlANVDofIksu3xpgoJdG4L7I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16UNManzuffvMTmsgfWnWF6Bt5jjvtS0tBsdToDQ_bZvc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 24832 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 8448 + LdsOffsetMetadata_Blk: 24832 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1kHn9o1wGnb0Le03rKuL4bApCYLkFlSLtoU6MjkRibzc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1NMWe8o4PG1BIiKP1ggEjZUpwkkJSDyvBBNMZ8tXiaR8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 1 + NumLoadsB: 7 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1Q0rm-ZJ-B1Ep3tDRV7Y6v5xFppIw1pc1wB6W7qCB6eE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 32000 + LdsInitCVgprs: false + LdsNumBytes: 32000 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32000 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI14_XY01zUqMSgVV3Z29zc0HvqY2RBU1XaMixpmKPTFvQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1BV068w-wQamiyACMJrCQO6AzvkFtVVaL6YUOIybq_As= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI167ogjNu2arPxWNgqgKM8sDxj_QbeHe5G3biyDU4kITPE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 19712 + LdsInitCVgprs: false + LdsNumBytes: 19712 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 19712 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI1606uPSrg8jSF_adB8-Q1EUFxmHwf5F9lLMz64s3UmyCk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16Jm6coo6HlXoYc23v3SkS10L-uM9lwa5bnHnY2Vs_oLo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIv9D6qCD557UuGj0WYoB1KLme1Sh1mzoR8zvP-Ew1rDY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 66560 + LdsInitCVgprs: false + LdsNumBytes: 66560 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 66560 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 13] + MIWaveTileA: 4 + MIWaveTileB: 13 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 208 + MacroTileA: 256 + MacroTileB: 208 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 52 + NumLoadsA: 8 + NumLoadsB: 13 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 13 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 13 + ThreadTileA: 16 + ThreadTileB: 13 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIZTzgVMYbF8zAjkLbWPx_KguNjRPGDWzQphgfwu106mI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 28160 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 11] + MIWaveTileA: 4 + MIWaveTileB: 11 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 176 + MacroTileA: 256 + MacroTileB: 176 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 44 + NumLoadsA: 8 + NumLoadsB: 11 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 11 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 11 + ThreadTileA: 16 + ThreadTileB: 11 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIYpXgEPF2KmPhp_HSItkaLdu8ZfTql6wuzRMctpM8bWs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 121856 + LdsInitCVgprs: false + LdsNumBytes: 121856 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 9] + MIWaveTileA: 4 + MIWaveTileB: 9 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 144 + MacroTileA: 256 + MacroTileB: 144 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 8 + NumLoadsB: 9 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 9 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 9 + ThreadTileA: 16 + ThreadTileB: 9 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIM6ky9Tq6VRezat1v-Vy4Qy4OEEUGyFJHwE-XvlIpRaY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 17920 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 112 + MacroTileA: 256 + MacroTileB: 112 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1ZaT28UTzCJBM6wCUWArJk9cu4afCdA6oGhULOQC7jwQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 46080 + LdsInitCVgprs: false + LdsNumBytes: 46080 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 12800 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46080 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1ckLt2U3Q4BSlwSpNR-cpd6uJ3sMr5OVPRWuKvEiSVtY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 7680 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1Cks6yoWu7mRHY5OU_L0wlyt_BkdBZ9B6KYLX0jMjEZM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 35840 + LdsInitCVgprs: false + LdsNumBytes: 35840 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 2560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1LRieWm8obTv5BrugK6ZfaH9J4-zT3LrRBcx6GxwPdUk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 1280 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45312 + LdsInitCVgprs: false + LdsNumBytes: 45312 + LdsNumElementsAlignedA: 10496 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 10496 + LdsOffsetB_Blk: 76032 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45312 + LdsOffsetMetadata_Blk: 76032 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1zBVp9O_j-pDQmFJ0pruVZjPjyXjbaxIDNKuOKzE-hUk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 768 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 6400 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 6400 + LdsOffsetB_Blk: 71936 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 71936 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1LmEbjkFjRe2EHvxZ5njVJSBpnw78dqd37eC05zoq8oY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 2304 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2304 + LdsOffsetB_Blk: 67840 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 67840 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI_Cpsqq7IdCcI6m09ixnRrSz_HTRJbwY41LkpuZ6eaow= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1tP0yLw5sOaqSqLw69T0rUISjIec4gtRcEBnLindXxd0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1WlKSkQYWc45q9-9u3Sa8KR5G5YSwdMXXvYIooja5mzU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1ZFz2Wqi9AGPfO9YZeT8hu0U-a6QRpBHZ1MNZhOuWfzs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16YTIDOkvWpHTvJMbXMK3Qn2TUcHpLDIz3L67zYqwLtxM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1Hyiuad9E_LwogHYOQgHONBkR83eXTUE_-HwHw1z6c2M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 62976 + LdsInitCVgprs: false + LdsNumBytes: 62976 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 62976 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 10 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1_vkhGjNt0PNt_xV8VVOz2XzUCfwEcU8zljVlbKTICVg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16riuLixhIaRRQRAFH27-FM-5iTnYrSUKGh2HFdVNH4W8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16tRLsgniRf9NQ8cwUknJpPeoosk2I0euxrG_yXuRD6Zc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16zCjLhPMzebcvjlA5OUrujTBGN1OYulB3ktj1a570vLs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1OuevmgTEfRQM80_r7hzorlLPNIZiSHCEk0LhjgcOAH8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 60672 + LdsInitCVgprs: false + LdsNumBytes: 60672 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 60672 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 12 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI117HfbCeMXp_75IhCNgyVsAtudVa47Y080duX4tlb_Cc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 54528 + LdsInitCVgprs: false + LdsNumBytes: 54528 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 54528 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 10 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1w_TVfNrjE3mYvq96Bm1Jgkl-HCbXvzNQ6H6gr4PSnKc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16IX89BCLJtYzlNb-bQG51jUiesgW7gn9abBGKzaC4u7Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16yEC0PHvmD0zrKvRDTkhqsmiXq78VvD0Eg5fcAnocoic= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16_OtB8fTrZk3Pt0bE3bXfsWfkoWXEPa2ENPfe4eaiVRI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1uaMQi1863lUOUemwTE99mttXxpTRkvqvmiwvtfx6Qx0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 80384 + LdsInitCVgprs: false + LdsNumBytes: 80384 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 13824 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 80384 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 16 + NumLoadsB: 3 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1_aftRJpk6u1tTs3-Hwls-Fo3A3dD889Os-hWuvfNQvw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 71168 + LdsInitCVgprs: false + LdsNumBytes: 71168 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 4608 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71168 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI18CjcSziK0PjBic2MAopWR-Ta_nHbDse6qNtds5pye7Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 256 + LSPB: 16 + LVCA: 1 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 71936 + LdsInitCVgprs: false + LdsNumBytes: 71936 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 135424 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71936 + LdsOffsetMetadata_Blk: 135424 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 512 + LSPA: 64 + LSPB: 8 + LVCA: 4 + LVCB: 32 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI169OaDNKWZnXMa1R_AlPEgQpVqAFVb54TBTn3PvUMxY7A= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 512 + LSPA: 128 + LSPB: 8 + LVCA: 2 + LVCB: 32 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16JRL6UviIw3ql2MWJlkjDhAPVUQuM-dfo_YP9V8INeCE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16a9rQBxUkyvDWlmJquyIGWqzMINaNpRD_aWL9QWIuzQ8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 512 + LSPA: 128 + LSPB: 8 + LVCA: 2 + LVCB: 32 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16Jyq1ax9kWbpfAk08jNUVf7z1XoXRVCVdpt-kBpa0fWY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 512 + LSPA: 64 + LSPB: 8 + LVCA: 4 + LVCB: 32 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16D0scrs9_FOYkVHUHqfaWIBeu8Rz72d3pOV3R8iV8H74= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI169Ao_eCjaajChoPx9i0y3MGqMyeJdbqU3dbjwB_UIKx4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1DDaFTcewwV5Ybn8swfgQtfH13aPb98dLOnRrTIJYmSQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1ifWE51xli667eSWHdf8X0Gqm1isofKMCZj08QiVl3U0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 1024 + LSPA: 128 + LSPB: 4 + LVCA: 2 + LVCB: 64 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1U06udU6pkFuvB0exyiBHs-wOgMdXOgbcv4nENk4uw44= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1oDw0RAqdNEFhP7idqV0O87totk9ugWGoiXKFb7JRQVw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84992 + LdsInitCVgprs: false + LdsNumBytes: 84992 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84992 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 16 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xqwn4mYe9PqjtRIapRznDsflXKMnaPO3v98RcshyCRoc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 16 + LdsPadB: 8 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index c5941f95691b..289cb40932c4 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,19809 +81,14277 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI93W5TiEccmFg8wGd3V3_i7anSx2in1g2Nvgr1nxFMW0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1WBDB2zkUie8-CEEoI_ArcQdbGzpndP9LCMEirfw4oOM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46336 - LdsInitCVgprs: false - LdsNumBytes: 46336 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46336 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1byiwSEFKpQOHfD3AL3aMNi54Gti0tPrWia0W_RI3F9I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1drE7ijfboWVMJx6XbcD7R5giTiA773IMYwA0SWdbSew= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIlwPgpsfEr1tHJhWc_dQ16xuUglNHBRvb9FY8kgrV2uo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 128 - MacroTileA: 224 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1vpzVkbJ5WB1luwHLUPLp6GT1RJDqsKYFBruB0CuPp3k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48384 - LdsInitCVgprs: false - LdsNumBytes: 48384 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48384 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 3] - MIWaveTileA: 7 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 96 - MacroTileA: 224 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 7 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 3 - ThreadTileA: 28 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1zuV98IyOwxElPoy5ichUD4fa6nssRrdqd1Q3modyHdk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44288 - LdsInitCVgprs: false - LdsNumBytes: 44288 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44288 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 7 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1xqlNtEYE-E0Mz0QATa9_dzumE9-a3Wvr1cR9tzlzYW8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 7 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICgefWmv3HWofIYKXJsq8855MHgkupzgrkVMcs4u-by0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 125952 - LdsInitCVgprs: false - LdsNumBytes: 125952 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 8] - MIWaveTileA: 6 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 256 - MacroTileA: 192 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 8 - ThreadTileA: 24 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MInHiNZ23NKfFsK7oWehXsdVhdC3keTzJy0tZ_ibW9pzc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1rAFgL4PTZQ2gx_CQR3J0YNYjnR6ZpZqLT3kKoVcsb8Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1Vg4WjQQgNmrK6Ytw7ZwOj3TBnXGNte5TZoyB4kGGvxQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1WRxklg9QBBuMW_MrlGv_hyeD_1KjPxvE5CvBb2xGGdk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32000 - LdsInitCVgprs: false - LdsNumBytes: 32000 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 60416 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32000 - LdsOffsetMetadata_Blk: 60416 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MIGmZq9BXDiHJ0tc5zxXOziePTEBPwz6cUZrAUknIRT7E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123904 - LdsInitCVgprs: false - LdsNumBytes: 123904 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 8] - MIWaveTileA: 5 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 256 - MacroTileA: 160 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 160 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 8 - ThreadTileA: 20 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIwjK4OAX-B9dLgo7aLraklJg0ipZLFjGXad70tFlHa_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50432 - LdsInitCVgprs: false - LdsNumBytes: 50432 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50432 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 6] - MIWaveTileA: 5 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 192 - MacroTileA: 160 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 120 - NumLoadsA: 5 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 6 - ThreadTileA: 20 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIXGlL2yJ1nYGKMcvOKU86eOyg9fj72cyU6dwaNXql1aE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1d_LfxY9vu-XLyyQyjblMOjGGccv6SWDTcAMP2nTGrZ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1xJMqskwq5v3yaRA45HS_l-Q0PeuwEvXnjkyc8iEsIuQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1J2YusILFQCwZJTPycybXWTPnKyMth_wvsZJGJUL6XBA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 58368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 58368 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIK3HX_ARTjdrJyQgn8ka1A08cR0NqiFPRF8k7JYx3lno= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50176 - LdsInitCVgprs: false - LdsNumBytes: 50176 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50176 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 8] - MIWaveTileA: 4 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 8 - ThreadTileA: 16 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIAYYyDwfBJqmmCRgY0wORDPvjgSZIA73oaclExRH_3Ks= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIL86z76qmkuTn8D76ItocU1Fqkm1UTu79IZpG8k_14Mo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MIcEWO-iBcRFoqXFm5FnHIJnGhh114ULGYJcbTf6WzdFI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1uj-yswGYT2k5VXIiR3VHictFIePTXmx0STeF7oDnJ6c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1_uypCl737WTznRBFG8qQSTk3KV2VA62EV3JnGAPhybs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1SN9OiW0wmLBhcMdgMSszjYfbUYfIwyxoyCZVswMyYtQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1OdZvyKg_z2Mm3twNkO33DsDFI4XHPhixfFGT8M7-fns= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48128 - LdsInitCVgprs: false - LdsNumBytes: 48128 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48128 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1-rYnfwe4wvrEHrwuVO2V79PzVjbfHc2PuGqRKVJDgfI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1QbCX-NbN9_FulNkEv2WcOQVCSkdBBRKzE6PDyeD0PIc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI15g6oi8kYRfTdSkq4hb7l4vHYA3S6a_b3gQ32PTizD-c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32256 - LdsInitCVgprs: false - LdsNumBytes: 32256 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32256 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16zaeoUJs0j0gHGLtiiAi_mKOg4oYgz97kp3_f949MqD0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 27904 - LdsInitCVgprs: false - LdsNumBytes: 27904 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27904 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16n1I05diAnU09ahBMg4ZOJ69DUU98umJt7phil649P_I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 23808 - LdsInitCVgprs: false - LdsNumBytes: 23808 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23808 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16qYVsRYavL1_uEB81mfehkE0xypBMpwpmv8NTKwWUCZQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 19712 - LdsInitCVgprs: false - LdsNumBytes: 19712 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 19712 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1rLkRPc5Fv6KcZ0pkpjti_dDNkwatHGJAjTt5YgSUF0A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1MsUtatzu8gT7lwtRqOR1GlkYA3LAi1h7-WpiR8njS-s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI17ixososrZhbGip5glmYhzY0dhoWQ6s92NzkMuYpNd6c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1O6oTiyV66Nmoj0i8-AQlJj5R-NLtNsfH6diaQrWbGzQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16BDy2huNg1pYAEVpwRmCSx6TZ3y-fyTBhxwPrXBTUMn0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16T76NxLk5ebc1y9bf0bGaRoLE1Btglvx1amB35nt2jrg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI164mp4kBtFpiXW7TfMmUnYxQd7fjnanNrZue8IY8R1Nt4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29824 - LdsInitCVgprs: false - LdsNumBytes: 29824 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 25600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 9216 - LdsOffsetMetadata_Blk: 25600 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1CBdovKQljtt9JJeN35QUWWaGXoZh_9PeLDfb3HuWMiA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37888 - LdsInitCVgprs: false - LdsNumBytes: 37888 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37888 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1eYnh8ExIG9i1WyAXPWhGKC2q4srWo55-2vcORDH9kEs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1m1tiKe28X_8BSOyU__5gEgsr8TQY2h1rX5gF4hmE4Qc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI17VRJH_5pnmfoUU-FHr32iGMlOdeQUYJplapfD41zqrY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22016 - LdsInitCVgprs: false - LdsNumBytes: 22016 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22016 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16AqtF9WEJ-EpQbhJnOZKbC372wZ4o2kh1rPtJZau1G1U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16WqqFFjCIevU2bTOwvq53GHxOVUs4xuB_iw0OyIMRO9s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16O0pSqsQ7I3lGvs23ES_yKR_wzzq3HSwM1IcyJmGtnhc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25728 - LdsInitCVgprs: false - LdsNumBytes: 25728 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1Jybduy0hX_yQp-cMS6QlHSWe9YueMgbIOjgYTfyE4xY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1280 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45312 - LdsInitCVgprs: false - LdsNumBytes: 45312 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 10496 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45312 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1Mhf-Vi17l2Fosir59CVL3Da6ElGCICe4ioDdRimtZBI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 768 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 6400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1zfbotELeA8YVwjd3HOC5VjWft_ZNaVArBeo8Zgzp6wo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 2304 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIPJDSEwAxzWM4fDuAEI3dit5Qm8ipFSRMSAIZnl6eve4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 66560 - LdsInitCVgprs: false - LdsNumBytes: 66560 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 66560 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [13, 4] - MIWaveTileA: 13 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 208 - MacroTile1: 256 - MacroTileA: 208 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 208 - NumLoadsA: 13 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 13 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 52 - ThreadTile1: 4 - ThreadTileA: 52 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIHlGj0Igh78VESckoEVhEEojHcEyAVYdCerN_cX6Ldrw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 28160 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 28160 - LdsOffsetB_Blk: 93696 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 28160 - LdsOffsetMetadata_Blk: 93696 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [11, 4] - MIWaveTileA: 11 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 176 - MacroTile1: 256 - MacroTileA: 176 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 176 - NumLoadsA: 11 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 11 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 44 - ThreadTile1: 4 - ThreadTileA: 44 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MIs43vjb8vXLGfeBgy4ohTkGSxXW8PsCp3uz_zeqpVxeM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 121856 - LdsInitCVgprs: false - LdsNumBytes: 121856 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 88576 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23040 - LdsOffsetMetadata_Blk: 88576 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [9, 4] - MIWaveTileA: 9 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 144 - MacroTile1: 256 - MacroTileA: 144 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 144 - NumLoadsA: 9 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 9 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 36 - ThreadTile1: 4 - ThreadTileA: 36 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIYY998OJmuRcBnYy6KGEItwDq2j2QehUOs8VfioGt_EY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17920 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17920 - LdsOffsetB_Blk: 83456 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 83456 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 112 - MacroTile1: 256 - MacroTileA: 112 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1riWq2NrSVGbHY4ulT3CBZ1ka9jZSPImqGDsOLfR739U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46080 - LdsInitCVgprs: false - LdsNumBytes: 46080 - LdsNumElementsAlignedA: 12800 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12800 - LdsOffsetB_Blk: 78336 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46080 - LdsOffsetMetadata_Blk: 78336 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI109tIWmrNFpIMCvCKD44FEjLK6mNrPP0ecMYRKJWv__o= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 7680 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 7680 - LdsOffsetB_Blk: 73216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 73216 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1ePlkOMYOjA6Sj3U5aK9R-vdkoYkQzZKOf49wW2bLE70= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35840 - LdsInitCVgprs: false - LdsNumBytes: 35840 - LdsNumElementsAlignedA: 2560 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2560 - LdsOffsetB_Blk: 68096 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 68096 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1R3pW5TGY8A5opfshY0CC3g5uD4DXiGRwS1VnQHSdtTs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 60672 - LdsInitCVgprs: false - LdsNumBytes: 60672 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 117760 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 60672 - LdsOffsetMetadata_Blk: 117760 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI13Sbf9PQEx5F-iKE9julI7_PdDt3yPapFP4vCFUpDo1E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 62976 - LdsInitCVgprs: false - LdsNumBytes: 62976 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 62976 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 10 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI137t5Uvc0ssl50sE4SNlCcUp-XUoCcLX9_I4uGMXledk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 54528 - LdsInitCVgprs: false - LdsNumBytes: 54528 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 54528 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 10 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIOmRHwGdcC8gFOr3dPzasFfQWs0PP2cDcwy6aKZU2f5U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1454Bwxd_MIvQjuG-Y_CvHs_qnR7PgRRvHGhCUO-XbOY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1Rmqg4myY3vDAMf_l0hT_13f8mLa__KusFSpIvX00Gq4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI15NEBMkxG3p9Ig0Cr9U7Vscm7bIb_z6qr2ylZCFd2y30= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16wxJkzA64nljgT6A4tMYPR-qnvL3NK7-LGLeTrL93K28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16s4XHH5Mm_pX3J5f5thK2zPFz4wk3wfVK1ZsVsKy_LpE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1XeU_kq-TCV7gFRA0zjadiPg5xVzK4J_9tcfhp6RHcE8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16uy_AuCnsXVT_W8ZUwPY2xZhbdvr0s1-k6m6FDS2GyPc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16JHWtGGP8udHdAtFC_ITt55YeNZjw7qy9YbaDQmBmLnw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1CMycMeJchCKewZBbLPkfdDttQ7jpxPNi2DcgEi0aA7Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI167PfjxLOUHYcmsWqM3Jr5QHvep_HsB981kR15bVmyR_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16hW4G8jnTFUiWZWnW0Tmfmi9zpySUXkWHngzRijqsswA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16pkwZPXEjbcu5SfItYZgOkhxquryDik7C3Mlw6DhdC9c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1ag0qOUjDoYcyoioR4OdNEh8L5HMCVTkf-1kVLG0Ke8Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 256 - LVCA: 16 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71936 - LdsInitCVgprs: false - LdsNumBytes: 71936 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71936 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1RElBQ1lA5tfxLjjMGZY4_t9WElkqvIrWw-ZvypHxsS4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 80384 - LdsInitCVgprs: false - LdsNumBytes: 80384 - LdsNumElementsAlignedA: 13824 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 13824 - LdsOffsetB_Blk: 144896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 80384 - LdsOffsetMetadata_Blk: 144896 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1g4O7XQYq5Erjhq0mbnd0ZzsOn7Y74NZ0Ed50NCU8Cko= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71168 - LdsInitCVgprs: false - LdsNumBytes: 71168 - LdsNumElementsAlignedA: 4608 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4608 - LdsOffsetB_Blk: 135680 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71168 - LdsOffsetMetadata_Blk: 135680 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 64 - LSPA: 8 - LSPB: 64 - LVCA: 32 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16c-3ymAqE5iIzAVIl4M4mFas_yKBzwB8S5hIJHrx_G10= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 32 - LSPA: 8 - LSPB: 128 - LVCA: 32 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16l2koiE9edxgF5EcPYxeXQhdktTUf3O_Nh5I8-rOv_5c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16PDTjUmNdBPeXIoutsSh54JIRkDgOgTuGJ5_UUnXYC30= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16FZJXxZzBUYpkHY1uHTOo5xfvmHSftJ7Ss8JFmsnHOq4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16UExxzleg69jKCOIPv3aHKRuxpeCJ0OtN8U8_i8380wE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 32 - LSPA: 8 - LSPB: 128 - LVCA: 32 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16486QKgKm2Q-ueCL9nw3616pMoSPYJwy96wpaKy2ZR9g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 64 - LSPA: 8 - LSPB: 64 - LVCA: 32 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1gJ-iaCReTZwgtOaq-r_7Ae1Xx5lpHHILXpRm4hXIFWo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1TOymj1f06yNxQYQNnDElBwV9Je3kDkNSQdL4ZFlGHxw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1eVyecsmYTDI_tnvIi0T4aak3HoYfyCs1_BMjsWhSVGA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84992 - LdsInitCVgprs: false - LdsNumBytes: 84992 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84992 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1SvhFCnFpOT5VXBaOtUT022uRF_6hTz1blD7UqHf2kpA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 32 - LSPA: 4 - LSPB: 128 - LVCA: 64 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xDWExf5YqS6e0I09q-b7LtSPBEclPiz1pvMgNxDpD5yY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 8 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI93W5TiEccmFg8wGd3V3_i7anSx2in1g2Nvgr1nxFMW0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1WBDB2zkUie8-CEEoI_ArcQdbGzpndP9LCMEirfw4oOM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 46336 + LdsInitCVgprs: false + LdsNumBytes: 46336 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46336 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1byiwSEFKpQOHfD3AL3aMNi54Gti0tPrWia0W_RI3F9I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1drE7ijfboWVMJx6XbcD7R5giTiA773IMYwA0SWdbSew= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIlwPgpsfEr1tHJhWc_dQ16xuUglNHBRvb9FY8kgrV2uo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 128 + MacroTileA: 224 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 4 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1vpzVkbJ5WB1luwHLUPLp6GT1RJDqsKYFBruB0CuPp3k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 48384 + LdsInitCVgprs: false + LdsNumBytes: 48384 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48384 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 3] + MIWaveTileA: 7 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 96 + MacroTileA: 224 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 7 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 3 + ThreadTileA: 28 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1zuV98IyOwxElPoy5ichUD4fa6nssRrdqd1Q3modyHdk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 44288 + LdsInitCVgprs: false + LdsNumBytes: 44288 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44288 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 7 + NumLoadsB: 2 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1xqlNtEYE-E0Mz0QATa9_dzumE9-a3Wvr1cR9tzlzYW8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 7 + NumLoadsB: 1 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICgefWmv3HWofIYKXJsq8855MHgkupzgrkVMcs4u-by0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 125952 + LdsInitCVgprs: false + LdsNumBytes: 125952 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 8] + MIWaveTileA: 6 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 256 + MacroTileA: 192 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 8 + ThreadTileA: 24 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MInHiNZ23NKfFsK7oWehXsdVhdC3keTzJy0tZ_ibW9pzc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1rAFgL4PTZQ2gx_CQR3J0YNYjnR6ZpZqLT3kKoVcsb8Y= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1Vg4WjQQgNmrK6Ytw7ZwOj3TBnXGNte5TZoyB4kGGvxQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1WRxklg9QBBuMW_MrlGv_hyeD_1KjPxvE5CvBb2xGGdk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 32000 + LdsInitCVgprs: false + LdsNumBytes: 32000 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 60416 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32000 + LdsOffsetMetadata_Blk: 60416 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MIGmZq9BXDiHJ0tc5zxXOziePTEBPwz6cUZrAUknIRT7E= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 123904 + LdsInitCVgprs: false + LdsNumBytes: 123904 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 8] + MIWaveTileA: 5 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 256 + MacroTileA: 160 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 160 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 8 + ThreadTileA: 20 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIwjK4OAX-B9dLgo7aLraklJg0ipZLFjGXad70tFlHa_4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 50432 + LdsInitCVgprs: false + LdsNumBytes: 50432 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50432 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 6] + MIWaveTileA: 5 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 192 + MacroTileA: 160 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 120 + NumLoadsA: 5 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 6 + ThreadTileA: 20 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIXGlL2yJ1nYGKMcvOKU86eOyg9fj72cyU6dwaNXql1aE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1d_LfxY9vu-XLyyQyjblMOjGGccv6SWDTcAMP2nTGrZ4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1xJMqskwq5v3yaRA45HS_l-Q0PeuwEvXnjkyc8iEsIuQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1J2YusILFQCwZJTPycybXWTPnKyMth_wvsZJGJUL6XBA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 58368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 58368 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIK3HX_ARTjdrJyQgn8ka1A08cR0NqiFPRF8k7JYx3lno= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 50176 + LdsInitCVgprs: false + LdsNumBytes: 50176 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50176 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 8] + MIWaveTileA: 4 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 8 + ThreadTileA: 16 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIAYYyDwfBJqmmCRgY0wORDPvjgSZIA73oaclExRH_3Ks= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIL86z76qmkuTn8D76ItocU1Fqkm1UTu79IZpG8k_14Mo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MIcEWO-iBcRFoqXFm5FnHIJnGhh114ULGYJcbTf6WzdFI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1uj-yswGYT2k5VXIiR3VHictFIePTXmx0STeF7oDnJ6c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1_uypCl737WTznRBFG8qQSTk3KV2VA62EV3JnGAPhybs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1SN9OiW0wmLBhcMdgMSszjYfbUYfIwyxoyCZVswMyYtQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1OdZvyKg_z2Mm3twNkO33DsDFI4XHPhixfFGT8M7-fns= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 48128 + LdsInitCVgprs: false + LdsNumBytes: 48128 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48128 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1-rYnfwe4wvrEHrwuVO2V79PzVjbfHc2PuGqRKVJDgfI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1QbCX-NbN9_FulNkEv2WcOQVCSkdBBRKzE6PDyeD0PIc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI15g6oi8kYRfTdSkq4hb7l4vHYA3S6a_b3gQ32PTizD-c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 32256 + LdsInitCVgprs: false + LdsNumBytes: 32256 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32256 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16zaeoUJs0j0gHGLtiiAi_mKOg4oYgz97kp3_f949MqD0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 27904 + LdsInitCVgprs: false + LdsNumBytes: 27904 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27904 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16n1I05diAnU09ahBMg4ZOJ69DUU98umJt7phil649P_I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 23808 + LdsInitCVgprs: false + LdsNumBytes: 23808 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23808 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16qYVsRYavL1_uEB81mfehkE0xypBMpwpmv8NTKwWUCZQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 19712 + LdsInitCVgprs: false + LdsNumBytes: 19712 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 19712 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1rLkRPc5Fv6KcZ0pkpjti_dDNkwatHGJAjTt5YgSUF0A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1MsUtatzu8gT7lwtRqOR1GlkYA3LAi1h7-WpiR8njS-s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI17ixososrZhbGip5glmYhzY0dhoWQ6s92NzkMuYpNd6c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1O6oTiyV66Nmoj0i8-AQlJj5R-NLtNsfH6diaQrWbGzQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16BDy2huNg1pYAEVpwRmCSx6TZ3y-fyTBhxwPrXBTUMn0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16T76NxLk5ebc1y9bf0bGaRoLE1Btglvx1amB35nt2jrg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI164mp4kBtFpiXW7TfMmUnYxQd7fjnanNrZue8IY8R1Nt4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29824 + LdsInitCVgprs: false + LdsNumBytes: 29824 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 25600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 9216 + LdsOffsetMetadata_Blk: 25600 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1CBdovKQljtt9JJeN35QUWWaGXoZh_9PeLDfb3HuWMiA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 37888 + LdsInitCVgprs: false + LdsNumBytes: 37888 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37888 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1eYnh8ExIG9i1WyAXPWhGKC2q4srWo55-2vcORDH9kEs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1m1tiKe28X_8BSOyU__5gEgsr8TQY2h1rX5gF4hmE4Qc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI17VRJH_5pnmfoUU-FHr32iGMlOdeQUYJplapfD41zqrY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 22016 + LdsInitCVgprs: false + LdsNumBytes: 22016 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22016 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16AqtF9WEJ-EpQbhJnOZKbC372wZ4o2kh1rPtJZau1G1U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16WqqFFjCIevU2bTOwvq53GHxOVUs4xuB_iw0OyIMRO9s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16O0pSqsQ7I3lGvs23ES_yKR_wzzq3HSwM1IcyJmGtnhc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25728 + LdsInitCVgprs: false + LdsNumBytes: 25728 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1Jybduy0hX_yQp-cMS6QlHSWe9YueMgbIOjgYTfyE4xY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1280 + LdsBytesNoAmax: 45312 + LdsInitCVgprs: false + LdsNumBytes: 45312 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 10496 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45312 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1Mhf-Vi17l2Fosir59CVL3Da6ElGCICe4ioDdRimtZBI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 768 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 6400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1zfbotELeA8YVwjd3HOC5VjWft_ZNaVArBeo8Zgzp6wo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 2304 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIPJDSEwAxzWM4fDuAEI3dit5Qm8ipFSRMSAIZnl6eve4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 66560 + LdsInitCVgprs: false + LdsNumBytes: 66560 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 66560 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [13, 4] + MIWaveTileA: 13 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 208 + MacroTile1: 256 + MacroTileA: 208 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 208 + NumLoadsA: 13 + NumLoadsB: 8 + NumLoadsPerpendicularA: 13 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 52 + ThreadTile1: 4 + ThreadTileA: 52 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIHlGj0Igh78VESckoEVhEEojHcEyAVYdCerN_cX6Ldrw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 28160 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 28160 + LdsOffsetB_Blk: 93696 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 28160 + LdsOffsetMetadata_Blk: 93696 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [11, 4] + MIWaveTileA: 11 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 176 + MacroTile1: 256 + MacroTileA: 176 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 176 + NumLoadsA: 11 + NumLoadsB: 8 + NumLoadsPerpendicularA: 11 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 44 + ThreadTile1: 4 + ThreadTileA: 44 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MIs43vjb8vXLGfeBgy4ohTkGSxXW8PsCp3uz_zeqpVxeM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 121856 + LdsInitCVgprs: false + LdsNumBytes: 121856 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 88576 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23040 + LdsOffsetMetadata_Blk: 88576 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [9, 4] + MIWaveTileA: 9 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 144 + MacroTile1: 256 + MacroTileA: 144 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 144 + NumLoadsA: 9 + NumLoadsB: 8 + NumLoadsPerpendicularA: 9 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 36 + ThreadTile1: 4 + ThreadTileA: 36 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIYY998OJmuRcBnYy6KGEItwDq2j2QehUOs8VfioGt_EY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17920 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17920 + LdsOffsetB_Blk: 83456 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 83456 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 112 + MacroTile1: 256 + MacroTileA: 112 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1riWq2NrSVGbHY4ulT3CBZ1ka9jZSPImqGDsOLfR739U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 46080 + LdsInitCVgprs: false + LdsNumBytes: 46080 + LdsNumElementsAlignedA: 12800 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12800 + LdsOffsetB_Blk: 78336 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46080 + LdsOffsetMetadata_Blk: 78336 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI109tIWmrNFpIMCvCKD44FEjLK6mNrPP0ecMYRKJWv__o= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 7680 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 7680 + LdsOffsetB_Blk: 73216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 73216 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1ePlkOMYOjA6Sj3U5aK9R-vdkoYkQzZKOf49wW2bLE70= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 35840 + LdsInitCVgprs: false + LdsNumBytes: 35840 + LdsNumElementsAlignedA: 2560 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2560 + LdsOffsetB_Blk: 68096 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 68096 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1R3pW5TGY8A5opfshY0CC3g5uD4DXiGRwS1VnQHSdtTs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 60672 + LdsInitCVgprs: false + LdsNumBytes: 60672 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 117760 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 60672 + LdsOffsetMetadata_Blk: 117760 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 2 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI13Sbf9PQEx5F-iKE9julI7_PdDt3yPapFP4vCFUpDo1E= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 62976 + LdsInitCVgprs: false + LdsNumBytes: 62976 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 62976 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 10 + NumLoadsB: 4 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI137t5Uvc0ssl50sE4SNlCcUp-XUoCcLX9_I4uGMXledk= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 54528 + LdsInitCVgprs: false + LdsNumBytes: 54528 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 54528 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 10 + NumLoadsB: 2 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIOmRHwGdcC8gFOr3dPzasFfQWs0PP2cDcwy6aKZU2f5U= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1454Bwxd_MIvQjuG-Y_CvHs_qnR7PgRRvHGhCUO-XbOY= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1Rmqg4myY3vDAMf_l0hT_13f8mLa__KusFSpIvX00Gq4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI15NEBMkxG3p9Ig0Cr9U7Vscm7bIb_z6qr2ylZCFd2y30= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16wxJkzA64nljgT6A4tMYPR-qnvL3NK7-LGLeTrL93K28= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16s4XHH5Mm_pX3J5f5thK2zPFz4wk3wfVK1ZsVsKy_LpE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1XeU_kq-TCV7gFRA0zjadiPg5xVzK4J_9tcfhp6RHcE8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16uy_AuCnsXVT_W8ZUwPY2xZhbdvr0s1-k6m6FDS2GyPc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16JHWtGGP8udHdAtFC_ITt55YeNZjw7qy9YbaDQmBmLnw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1CMycMeJchCKewZBbLPkfdDttQ7jpxPNi2DcgEi0aA7Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI167PfjxLOUHYcmsWqM3Jr5QHvep_HsB981kR15bVmyR_4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16hW4G8jnTFUiWZWnW0Tmfmi9zpySUXkWHngzRijqsswA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16pkwZPXEjbcu5SfItYZgOkhxquryDik7C3Mlw6DhdC9c= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1ag0qOUjDoYcyoioR4OdNEh8L5HMCVTkf-1kVLG0Ke8Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 256 + LVCA: 16 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 71936 + LdsInitCVgprs: false + LdsNumBytes: 71936 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71936 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1RElBQ1lA5tfxLjjMGZY4_t9WElkqvIrWw-ZvypHxsS4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 80384 + LdsInitCVgprs: false + LdsNumBytes: 80384 + LdsNumElementsAlignedA: 13824 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 13824 + LdsOffsetB_Blk: 144896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 80384 + LdsOffsetMetadata_Blk: 144896 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 16 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1g4O7XQYq5Erjhq0mbnd0ZzsOn7Y74NZ0Ed50NCU8Cko= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 71168 + LdsInitCVgprs: false + LdsNumBytes: 71168 + LdsNumElementsAlignedA: 4608 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4608 + LdsOffsetB_Blk: 135680 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71168 + LdsOffsetMetadata_Blk: 135680 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 64 + LSPA: 8 + LSPB: 64 + LVCA: 32 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16c-3ymAqE5iIzAVIl4M4mFas_yKBzwB8S5hIJHrx_G10= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 32 + LSPA: 8 + LSPB: 128 + LVCA: 32 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16l2koiE9edxgF5EcPYxeXQhdktTUf3O_Nh5I8-rOv_5c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16PDTjUmNdBPeXIoutsSh54JIRkDgOgTuGJ5_UUnXYC30= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16FZJXxZzBUYpkHY1uHTOo5xfvmHSftJ7Ss8JFmsnHOq4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16UExxzleg69jKCOIPv3aHKRuxpeCJ0OtN8U8_i8380wE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 32 + LSPA: 8 + LSPB: 128 + LVCA: 32 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16486QKgKm2Q-ueCL9nw3616pMoSPYJwy96wpaKy2ZR9g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 64 + LSPA: 8 + LSPB: 64 + LVCA: 32 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1gJ-iaCReTZwgtOaq-r_7Ae1Xx5lpHHILXpRm4hXIFWo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1TOymj1f06yNxQYQNnDElBwV9Je3kDkNSQdL4ZFlGHxw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1eVyecsmYTDI_tnvIi0T4aak3HoYfyCs1_BMjsWhSVGA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 84992 + LdsInitCVgprs: false + LdsNumBytes: 84992 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84992 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 4 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1SvhFCnFpOT5VXBaOtUT022uRF_6hTz1blD7UqHf2kpA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 32 + LSPA: 4 + LSPB: 128 + LVCA: 64 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xDWExf5YqS6e0I09q-b7LtSPBEclPiz1pvMgNxDpD5yY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 8 + LdsPadB: 16 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index d82e5ebd0506..f7973f6bca5f 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_nta4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,31283 +81,22382 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MIrZs5mdCNmpikVNpV2YQQNFBCAdivVS8D1AsBXrhUP4k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 8] - MIWaveTileA: 8 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 256 - MacroTileA: 256 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 256 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 8 - ThreadTileA: 32 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MIADOZHpI0e4kz3PekZsq-nbyBZU7EZIxEg7KcvUC6g3g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 7] - MIWaveTileA: 8 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 224 - MacroTileA: 256 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 224 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 7 - ThreadTileA: 32 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIszVB7i9sCthoswarzra0qLsMXdGUHQA9HKGzu-1z5Kk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 6] - MIWaveTileA: 8 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 192 - MacroTileA: 256 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 6 - ThreadTileA: 32 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI4oZDINgtVU-f19tz2HYWh_LmdGFiZplMuFYBiLKIRVE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 124928 - LdsInitCVgprs: false - LdsNumBytes: 124928 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 5] - MIWaveTileA: 8 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 160 - MacroTileA: 256 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 5 - ThreadTileA: 32 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIQveu_FRPQ7kB2acKv85tmm0lWhIAYF1e-bVgSb8HhRU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1nNTR_ZcKmsj4A34uOMZTZX_oilYaQuaQFL2SNhs8cXo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 49152 - LdsInitCVgprs: false - LdsNumBytes: 49152 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 49152 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI17E3Elp9R2iGRIjoSUJt7ENU0IkIEDCYD27G5goO_zfQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI14oL9i2wAQa7K6yEREmecBQB_z40_9_5z5qamIYTi9BA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38912 - LdsInitCVgprs: false - LdsNumBytes: 38912 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38912 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MIVqWf-zHb4fGrqaI_PNfaMIuBSbf24WaOtDVKEmRwTy0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 166912 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 166912 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 8] - MIWaveTileA: 7 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 256 - MacroTileA: 224 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 224 - NumGlobalWriteVectorsPerThread: 224 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 8 - ThreadTileA: 28 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI2FE1_xdi6gisBeCfrvgksfzzbx9Yx6-LxacJqFoorOs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71680 - LdsInitCVgprs: false - LdsNumBytes: 71680 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 166912 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71680 - LdsOffsetMetadata_Blk: 166912 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 7] - MIWaveTileA: 7 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 224 - MacroTileA: 224 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 196 - NumGlobalWriteVectorsPerThread: 196 - NumLoadsA: 7 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 7 - ThreadTileA: 28 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MItgCKUscKBBiMtBOmdwUt6qAHeM_MR7YbKWEkh39p8UQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 129024 - LdsInitCVgprs: false - LdsNumBytes: 129024 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 6] - MIWaveTileA: 7 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 192 - MacroTileA: 224 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 168 - NumGlobalWriteVectorsPerThread: 168 - NumLoadsA: 7 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 6 - ThreadTileA: 28 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI_UOU6HZrGnyaniPidkPyA2vc3eMgOLn6XHhdMyr3gok= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 5] - MIWaveTileA: 7 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 160 - MacroTileA: 224 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 140 - NumGlobalWriteVectorsPerThread: 140 - NumLoadsA: 7 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 5 - ThreadTileA: 28 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIdhuZzZrmslv6fIdikHYnhn1dpXNXnkfYYXLJoFZzt6s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 128 - MacroTileA: 224 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1b5RoEaxogDkty9pH0a4h4iqEK74VHsxBERA-H-NeUNY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 3] - MIWaveTileA: 7 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 96 - MacroTileA: 224 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 7 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 3 - ThreadTileA: 28 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1g3lxPhwHBbhj-5_vMz-ezv_Gct2IZefQGw3UsfUuGuM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 7 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1pRHgvlAR7N2hth7GOj12UHkPW1O97Tnl14UAnpJSv9c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 7 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICt7le9q16ngZmPo_ZNYkoWOjOg2jVjonbKgLeTEzLxQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 8] - MIWaveTileA: 6 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 256 - MacroTileA: 192 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 8 - ThreadTileA: 24 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MIFCEugnbAE1o_KNOxQTjmgk1-riRp7-bqr1XYLHQkDJI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 129024 - LdsInitCVgprs: false - LdsNumBytes: 129024 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 7] - MIWaveTileA: 6 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 224 - MacroTileA: 192 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 168 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 6 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 7 - ThreadTileA: 24 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MIsXAzSnuik23-u_Tqj9IA9yIVlTTHvUnKVAh5tdPfE-0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 120832 - LdsInitCVgprs: false - LdsNumBytes: 120832 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 6] - MIWaveTileA: 6 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 192 - MacroTileA: 192 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 6 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 6 - ThreadTileA: 24 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIdawCsxOV8Q1yfZkDghTj5ROObRZEHHk5r8TLSJ8-4DI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 5] - MIWaveTileA: 6 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 160 - MacroTileA: 192 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 5 - ThreadTileA: 24 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI96PVzcjbGKsmGAPHOizb5QovZZPMnBXqYWXYWCMR6RE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1XF9bGkyS2smPrKpZKRvXqZE_IlG80gFUg26F5mTZLis= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1cyGRGs5fNOjzNULtz2NEdDFcpvgD-53w5wck9Mz0rEA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1dXOVrMlvBwO6y7ESILkUB5p9GtB-Km9ykpVzg9YqhsY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 60416 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 60416 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI_FMKfIaIlw04mnTidFqT-Fnj0gDbPBKMLBy9AxGoipc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 124928 - LdsInitCVgprs: false - LdsNumBytes: 124928 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 8] - MIWaveTileA: 5 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 256 - MacroTileA: 160 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 160 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 8 - ThreadTileA: 20 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MIgvcBHckHBXx7rYEXvd58PMUeCwZSGsDJtMgKSLGOngQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 7] - MIWaveTileA: 5 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 224 - MacroTileA: 160 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 140 - NumGlobalWriteVectorsPerThread: 140 - NumLoadsA: 5 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 7 - ThreadTileA: 20 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIiC0cBFJQlkymU09JdEq7dUq1wvcSX9wghKOSQxkOErI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 6] - MIWaveTileA: 5 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 192 - MacroTileA: 160 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 120 - NumLoadsA: 5 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 6 - ThreadTileA: 20 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MIwREMu9dXpHu0hNi1Te8Ryumj0HYpn1BFkusQoNfhAQ8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 5] - MIWaveTileA: 5 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 160 - MacroTileA: 160 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 100 - NumGlobalWriteVectorsPerThread: 100 - NumLoadsA: 5 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 5 - ThreadTileA: 20 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIcrglZoxFbHHjwhvMHFjwAoFdYaccIqWMvEsbqy0yalI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1BUsuevlFp3qE9qBkCpl8ZHtyZp2nTMvZX1nz1cQRT7M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1IVM_kq6fQOztyL87pw0KKbmpiglUkaPXmejIGutUrh4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1bgH1ukcWIV-QEjZUrEMgbOOF-CIy8Yu_mpQb3L9Zq3g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 58368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 58368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51712 - LdsInitCVgprs: false - LdsNumBytes: 51712 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51712 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [32, 32, 64, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 64 - MatrixInstM: 32 - MatrixInstN: 32 - MatrixInstruction: [32, 32, 64, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 4 - NonTemporalD: 4 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 16 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU8_SUM1_SUS256_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM8_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 8 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: true - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 4 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 2 - SubGroup1: 128 - SubGroupA: 2 - SubGroupB: 128 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 64 - ThreadTile1: 2 - ThreadTileA: 64 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 1 - enableLDSTrA: 0 - enableLDSTrB: 0 - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIHAxeqAl5GZyYpWyj9YBlHLiAot0B1CVx9Nf_5OHRC1Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 224 - MacroTileA: 128 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIIlrnP4n1WnSi3fe0xoxepRfTuuLMMp-jgbnYdGnnvYE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI5bZ7yznefmkxpzzbB-XiEKlKRJu2AMeEgvXx60Tq5L0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI72iEPgzvyM2ZrWa9mJE7BjRqJtpIrgM81I_tGNEjK0c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1xf_ukgQ-Z-qNS4jmkabhCAr3mB7WAQFBVo0jxB_s6wM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1j8uVA7i_pGx8ZCBrMR8ZhK3U6bp7J7YjtbYnnKdPxDI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1M6uR2wC2h6sr2Hl9YrF8zhWGQ7RgoVsQ-TM7lCE6pPI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22528 - LdsInitCVgprs: false - LdsNumBytes: 22528 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22528 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1lw4esJJcKXjrN7F_gKTEASvr4FPioQKVPOZImphrvBY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 49152 - LdsInitCVgprs: false - LdsNumBytes: 49152 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 49152 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1JixbiZckG_HlXVPjjWZHNInPUwLf2xyuU7BTFa5yHI0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 7] - MIWaveTileA: 3 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 224 - MacroTileA: 96 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 3 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 7 - ThreadTileA: 12 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1wIqbeUYowBvH47eY0D90voHkVaoBg2lPdbX2d68nIJ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1J-2M-3q4LSCntZjUh3v8nZH3qUcDC89rCJ0Fdq3iMeI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1Vr1CG6o8iRf8BbhZJxSu0FVBXoX31LtlPuM8cs_YEXU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16Cnahg_B3KmoGxZkTO1OYZLBTHOrwHNY1VQd2t5IrylI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16W4yMmcRT_Rcp3ZNo2gSW3q4SJdd9i9DRRKcgqox59VE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24576 - LdsInitCVgprs: false - LdsNumBytes: 24576 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 24576 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16zNms4_iu6xgFQiq42WUbXrNnFNyslp04sbAKWAlTbog= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20480 - LdsInitCVgprs: false - LdsNumBytes: 20480 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20480 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI11aFcbk23grO4-hK8aBkA0Q2TJ_RzSnSefhCsshDFWE8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1BIQ9uHrC9E8FHJ9mzReo0log_q-mFe47Gvfw9lRTxKU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1oWIDEj_bU2sIdW6WZSqKJI8Okx_mH0MwWvgKg4p0Wh0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1DfS6ucItLGFYy3q0Zz-JZd7icG_ycHgazunawsT8Pns= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1jK-MUAAqzxLFJYDxWtp_lRnOH6G_Sa3LS7gsDkzJ5ZI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16qBSTIiXAL1b2k0UyETAi0pu4GSfWwKXX5o-ybzBtncE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24576 - LdsInitCVgprs: false - LdsNumBytes: 24576 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 24576 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16mOGETtdY_W1srxs4Mfgd04W-cV79aCR6raMD3HA6jWU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 18432 - LdsInitCVgprs: false - LdsNumBytes: 18432 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 18432 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16wIclDOa00a_BcP3sOUDpDc_qFY9RvbDE777DV1a9O7k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 25600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 9216 - LdsOffsetMetadata_Blk: 25600 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1pg1aJvOuqnmTxe_-2e9_1S4uFGoAaPhgR-zs28ui024= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38912 - LdsInitCVgprs: false - LdsNumBytes: 38912 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38912 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1bJJ5iziycwP81Vg_rFbnRobq-WE2Gw-XuD7kvGV58LU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 1 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1c60kFRwy-kwCbpVB3Be-Tp-X0igfgGU9wMtdCOAL5xM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI13_eugo-sWMKarMTBe1WI4hmzJdiGvToykcHBbj4WXow= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1gSIhM4a60TQQmpGca3zPH_1C9YMOrBSb2Bu9JrxvD-k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22528 - LdsInitCVgprs: false - LdsNumBytes: 22528 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22528 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16YH7g6N9l5w5imlP3Lp9mdDXwtO_ZtzNO9yUXMFVOjKQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20480 - LdsInitCVgprs: false - LdsNumBytes: 20480 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20480 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16IFZMDGDs3-q6koIU2xIB_XUAfzhmng9Mg_46GO3Rqf4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16rLKaVqye5ZVlglTUO9ioHrZ6WB-GJ_Qs2RqP4S8ZIXE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MIyoGz0G9DWmdNz6WjS2Bq6rCsCrR_mXM4td8eMkMYEDw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73216 - LdsInitCVgprs: false - LdsNumBytes: 73216 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 38400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 165888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73216 - LdsOffsetMetadata_Blk: 165888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 15] - MIWaveTileA: 4 - MIWaveTileB: 15 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 240 - MacroTileA: 256 - MacroTileB: 240 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 240 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 8 - NumLoadsB: 15 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 15 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 15 - ThreadTileA: 16 - ThreadTileB: 15 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIBnc7kWhy6AvjwbNc5adjziCfJiLAVAD1RzoNcS6OMZQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 68096 - LdsInitCVgprs: false - LdsNumBytes: 68096 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 165888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 68096 - LdsOffsetMetadata_Blk: 165888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 13] - MIWaveTileA: 4 - MIWaveTileB: 13 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 208 - MacroTileA: 256 - MacroTileB: 208 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 52 - NumLoadsA: 8 - NumLoadsB: 13 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 13 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 13 - ThreadTileA: 16 - ThreadTileB: 13 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIF5A-1D1ZIPjJsJJ7ttt4AlfEFPQG0riur7Zv_qXG5YI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 128512 - LdsInitCVgprs: false - LdsNumBytes: 128512 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 28160 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 11] - MIWaveTileA: 4 - MIWaveTileB: 11 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 176 - MacroTileA: 256 - MacroTileB: 176 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 44 - NumLoadsA: 8 - NumLoadsB: 11 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 11 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 11 - ThreadTileA: 16 - ThreadTileB: 11 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIyRCs1CYgk6qoLPVzZz2698DWR4wMJ7whNVJ8JwJfnZc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123392 - LdsInitCVgprs: false - LdsNumBytes: 123392 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 9] - MIWaveTileA: 4 - MIWaveTileB: 9 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 144 - MacroTileA: 256 - MacroTileB: 144 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 8 - NumLoadsB: 9 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 9 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 9 - ThreadTileA: 16 - ThreadTileB: 9 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIaVDafPaNDXaLtu56b1YkZsQUI1aLmxb3jgdu_dv65EA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 17920 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 112 - MacroTileA: 256 - MacroTileB: 112 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1jNVYmoC6fjSjbIADTf0f9l0Cs4TKifqjd6Xql0ehllk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 47616 - LdsInitCVgprs: false - LdsNumBytes: 47616 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 12800 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 47616 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1eX_SljyDEcS5rCPMhk2caJwb3hhe7qqzmb_HRJ0i-U0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 7680 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1fPZ5I9MI2WZKscvBb8L9C1UGdH6oze92UNxsd_rmjZ0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37376 - LdsInitCVgprs: false - LdsNumBytes: 37376 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 2560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37376 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI17whl_3k4Y1EhiqZzn7y-o-UTqlat-Sd3AsiBr6V3nc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73216 - LdsInitCVgprs: false - LdsNumBytes: 73216 - LdsNumElementsAlignedA: 38400 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 38400 - LdsOffsetB_Blk: 169472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73216 - LdsOffsetMetadata_Blk: 169472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [15, 4] - MIWaveTileA: 15 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 240 - MacroTile1: 256 - MacroTileA: 240 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 240 - NumGlobalWriteVectorsPerThread: 240 - NumLoadsA: 15 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 15 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 60 - ThreadTile1: 4 - ThreadTileA: 60 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIjK835T0h-3Ar1N2b4kmmt65KtYLTJhfaZH3LmR2auFA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 68096 - LdsInitCVgprs: false - LdsNumBytes: 68096 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 68096 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [13, 4] - MIWaveTileA: 13 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 208 - MacroTile1: 256 - MacroTileA: 208 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 208 - NumLoadsA: 13 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 13 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 52 - ThreadTile1: 4 - ThreadTileA: 52 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIgsJynaoGp5Lr3OC9AUIiNlgAW9kkq7qPUmgK4ohOA_k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 128512 - LdsInitCVgprs: false - LdsNumBytes: 128512 - LdsNumElementsAlignedA: 28160 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 28160 - LdsOffsetB_Blk: 93696 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 28160 - LdsOffsetMetadata_Blk: 93696 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [11, 4] - MIWaveTileA: 11 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 176 - MacroTile1: 256 - MacroTileA: 176 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 176 - NumLoadsA: 11 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 11 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 44 - ThreadTile1: 4 - ThreadTileA: 44 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI6ZUiH-qsm4LTRuwyb7uw6oTXUaPI3Vjpt7WHaUbBOfI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123392 - LdsInitCVgprs: false - LdsNumBytes: 123392 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 88576 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23040 - LdsOffsetMetadata_Blk: 88576 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [9, 4] - MIWaveTileA: 9 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 144 - MacroTile1: 256 - MacroTileA: 144 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 144 - NumLoadsA: 9 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 9 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 36 - ThreadTile1: 4 - ThreadTileA: 36 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIiu6Dap7c-mOezayg2ePWEWJWIUNWwq_GUAWuc6asW8k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 17920 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17920 - LdsOffsetB_Blk: 83456 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 83456 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 112 - MacroTile1: 256 - MacroTileA: 112 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1fyJI8fnAEv3sOPHcVPqOC2AU9kSqOftCPxastr6HZJE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 47616 - LdsInitCVgprs: false - LdsNumBytes: 47616 - LdsNumElementsAlignedA: 12800 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12800 - LdsOffsetB_Blk: 78336 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 47616 - LdsOffsetMetadata_Blk: 78336 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-Zo3r6x3T98ZMWTZ4H7LqLtk7xXmH_6zOjhPN09FdTk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 7680 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 7680 - LdsOffsetB_Blk: 73216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 73216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1wehOWUghk2A68IMMHO37hmuhc-kFrA6X9yfb9WRqhBg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37376 - LdsInitCVgprs: false - LdsNumBytes: 37376 - LdsNumElementsAlignedA: 2560 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2560 - LdsOffsetB_Blk: 68096 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37376 - LdsOffsetMetadata_Blk: 68096 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI1vCzuwEprjPexlyglCW3zRXsjX9xUmGeZzokhH60Z9yE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 75776 - LdsInitCVgprs: false - LdsNumBytes: 75776 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 75776 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI1SiLGOArtDjYGdvuIvIxglRBR2PASF7znHipyq968J9k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81920 - LdsInitCVgprs: false - LdsNumBytes: 81920 - LdsNumElementsAlignedA: 64512 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 64512 - LdsOffsetB_Blk: 195584 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81920 - LdsOffsetMetadata_Blk: 195584 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 14 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 14 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI1Atkrqghs5hZ7rFrRDMn1qRHk9P7RjRIX2jTHJaCZcIs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 64512 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 64512 - LdsOffsetB_Blk: 195584 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 195584 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 14 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 14 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI1H6kBMzYhDsiXY1vBEU2Ud8UhLYehCYuWeQaCulrUTHo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 12 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI153vNbJPjGrVniCPDOzS5UGzSK6NOLhTT3dYrLoo1YJ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 12 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1y0BaulKQmTtKr2XSp5P3dnaOK0k5UYUxfkhT1HxC_rU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 117760 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 117760 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MIUYoy0JGLYkrhIavz9gjog-Elk__Z0jYq-QXQEcN2PaA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 177152 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 177152 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 10 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI1cUVPZBoZ-mtmE8sTWF_Az_FU-gB_6I8zhwfiN1pEaks= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 177152 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 177152 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 10 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI1EM3GcfaRjwnLLnXwjtnf8XZHB699buVut-TXk4ZjWyc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 63488 - LdsInitCVgprs: false - LdsNumBytes: 63488 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 63488 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 10 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 88 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI15N7cdBjXALKtpaYkT17ouFsMqBdqYm1rqUc2cC7GeLE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 10 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 89 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MIU14cbwZ_t9NC-mBYr7N5MljJmeoGwBb-WUmjyy7ILhU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 90 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIM08mmLtqT14x3TxGefabTcO290EciaeH6GnjDeU8Bgo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 91 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1OG3k0SIeaxxBYmDDXCQJ0v824MXDB2cvmAH14Io8mlM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 92 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1Ee_Jq2fBrVEHXaPp-qikzlaFXpjSm2wKDPvD-MKxLM0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 93 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI13IeiqjCNSTMel5JBLjnHPiB9PE_JYWIIfXqePFminMc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 94 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI1qUn-crPudL6dXZ6rCcKnkE9s4w-RcuNNqOhYdbIrR28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 158720 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 158720 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 6 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 95 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI18wp9zZc9oencMlWPaDY7BUc_iQgMxwF-RQOzHet7J3I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 158720 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 158720 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 96 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI1LrtbILHycv46qRt6NxJYDFm4zT_v-o9V1TAKrHNeAeQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 97 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI164Fx4NhdfXwLWoN6X-Afen5a5uy0a8LeKZgpjvani3CM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 98 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16_Nc2lY4fwUC-cbMivJ1U_eNf_9vhMW0WYu6jNMKBzYI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 99 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16X2E4Wx9NVLCyWp4uy3bGT7QFxR20eVc2o2HQJqwi8Hk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 100 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI1ehl3l0UKCFUW0CqlYWmD5Gv6DpebCzV_7KPnbR_5y4Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81920 - LdsInitCVgprs: false - LdsNumBytes: 81920 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 64512 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81920 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 14 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 14 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 101 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI1PpwSxkO--tNKZnLh5u1Zqztz5GRs2qwDUtmWoQ32_0M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 102 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1BIcxM2OYKG0Jqn5GRpKyHvCNhj2aQceoHMvTm7lJo6Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 63488 - LdsInitCVgprs: false - LdsNumBytes: 63488 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 63488 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 103 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1-eL7HedP-uoMH-4uLF-vatbGDwtmp3GZhw8IuLzDHvE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 104 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16Tidk1Ou4UA72T_vnHnWAsymx18z-7WG5S7yV6EMzxAE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 105 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16P7nrnDgCqsPXJVFvUTHGSoy-i5STapWZDg4F8KHN2a8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 106 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI164FNCErQfeGGweRYwOlRLTlQQw-uihsMini7VG5MjqUs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 107 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI1Weg33rA_lR82ctIL5Wv-agG74LvDUGtagD8ZNXCm6m8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 75776 - LdsInitCVgprs: false - LdsNumBytes: 75776 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 140288 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 75776 - LdsOffsetMetadata_Blk: 140288 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 108 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI1yjWOfHPT7EqX9rhhXejY8_eiJ07brJZxCVBlCXR9DBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 64512 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 140288 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 140288 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 14 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 14 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 109 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1pVP6pjfyPWPVeerMg8FY1EyiGjH4B_j7tc4lRUzRPvg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 110 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI1q7mTED5UvVJrOPDvYFNPflDjUkX1okY5IJT3AZBddMM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 111 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1fFpYWYjPWsmrBYgoz9CIhz7OHVoysPvt9SgTel6GPWI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 112 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16mu07e3PYZ_jp2LQxX0lmvu4pmsw_265fxGpIWycVSjg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 113 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI162x9lNzvJGF2jhOPNNRp1Q3mFaUkUkcNWlM7ecLmXXfs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 114 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16-01AyvDsXfpTy-wvoK3K25-ZRKuFeCLIRsj5lVXzX_s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 18432 - LdsInitCVgprs: false - LdsNumBytes: 18432 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 18432 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 115 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI1LDs52gy7NI4WuOMQ_idbNXBwBK86orsq1YVyOo4IvKs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 90624 - LdsInitCVgprs: false - LdsNumBytes: 90624 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 90624 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 16 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 116 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1NztuuZzeaShCMYTqz64dRhvSvIEetpxjBMHS12EujLs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81408 - LdsInitCVgprs: false - LdsNumBytes: 81408 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 13824 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81408 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 16 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 117 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1-eS3GREuZbaF_WcdTrJ8tJN1VRBcjitHDgD1jEhWCcE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 72192 - LdsInitCVgprs: false - LdsNumBytes: 72192 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 4608 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 72192 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 118 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI1ZLUtUYYiOHV_3Bx1CweJ56xkOj29nbS2TUKPiIsgWeM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 90624 - LdsInitCVgprs: false - LdsNumBytes: 90624 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 154112 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 90624 - LdsOffsetMetadata_Blk: 154112 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 119 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1QlOH2GDodFrn6A0rxCKHrFBaU_QWfv580nm9xWdZY0k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81408 - LdsInitCVgprs: false - LdsNumBytes: 81408 - LdsNumElementsAlignedA: 13824 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 13824 - LdsOffsetB_Blk: 144896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81408 - LdsOffsetMetadata_Blk: 144896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 120 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1PZ1TyY_JO23lBPKX9VhACTOX37NNgYiBjf2A-tanW7w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 72192 - LdsInitCVgprs: false - LdsNumBytes: 72192 - LdsNumElementsAlignedA: 4608 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4608 - LdsOffsetB_Blk: 135680 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 72192 - LdsOffsetMetadata_Blk: 135680 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 121 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16Z6LYIKVcBKQJUXMfCTkH2UmyhUWm90vrVgFH5F5h_is= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 122 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16Gx0c38MGhJUAjADEf6sgUUCYEnSxT0HvEevn-TLzDBA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 123 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 124 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16tO84-u8CmDozhzOYFPjZx1aAnLl_RdoGaIlGdosGUzM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 125 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 126 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16nxXVLq-_LhX2N8-ggvlvGFb2CSQPdL0Jk31-3Yn8K8w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 127 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI1Uh3YqlpEBhxKyjcdbTJSpHtEarJLrJfEIa0Mnxwsb5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 8 - LoopUnroll: 1024 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 128 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16t0-mPljiEZtupRhWAFpnxBDrJen-FsvDD34Q7PN3s_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 129 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16CAJPr2P6olRKJqA2Fmiymyoh6JhX7nYYGS7ffsGJvmw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 130 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16K_Na50sAdM91rhXevAlyiM9lrmobRF_PSYV6N76rQzk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 131 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16dVDH5F1pzH8NV-uymOAiVxprnNui5yf2Mp5t1_LqLTs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 132 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16Uv-d7icfFxrZ-YUiRaTBsX3xXbc3HcmSMlEdtwh8Yb0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 133 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1TNTLRLqQ0uhz1zq5XPt3rxttGE1MT5QoFsMWFmpwZGg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 134 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI13HHr0kFgQAqJ2TjtZ8tTohe6Udz-5u0CwkMq-_9renQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 135 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1KAql7WR7THsIAKFHzmlcHQVpFofe0Vzt2xFE0wiTOfU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84480 - LdsInitCVgprs: false - LdsNumBytes: 84480 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84480 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 136 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1-LGVgPAQHxTiHAOxIZA3A-LLqgvwFhsRG4LoerQvYqg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 137 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1uwD3q8H7ROxJTYYpP8cvtJiz3DeoGvbO5IiJaZpCFnE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84480 - LdsInitCVgprs: false - LdsNumBytes: 84480 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 147968 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84480 - LdsOffsetMetadata_Blk: 147968 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 4 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 138 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MIrZs5mdCNmpikVNpV2YQQNFBCAdivVS8D1AsBXrhUP4k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 8] + MIWaveTileA: 8 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 256 + MacroTileA: 256 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 256 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 8 + ThreadTileA: 32 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MIADOZHpI0e4kz3PekZsq-nbyBZU7EZIxEg7KcvUC6g3g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 7] + MIWaveTileA: 8 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 224 + MacroTileA: 256 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 224 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 7 + ThreadTileA: 32 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIszVB7i9sCthoswarzra0qLsMXdGUHQA9HKGzu-1z5Kk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 6] + MIWaveTileA: 8 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 192 + MacroTileA: 256 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 6 + ThreadTileA: 32 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI4oZDINgtVU-f19tz2HYWh_LmdGFiZplMuFYBiLKIRVE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 124928 + LdsInitCVgprs: false + LdsNumBytes: 124928 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 5] + MIWaveTileA: 8 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 160 + MacroTileA: 256 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 5 + ThreadTileA: 32 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIQveu_FRPQ7kB2acKv85tmm0lWhIAYF1e-bVgSb8HhRU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1nNTR_ZcKmsj4A34uOMZTZX_oilYaQuaQFL2SNhs8cXo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 49152 + LdsInitCVgprs: false + LdsNumBytes: 49152 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 49152 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI17E3Elp9R2iGRIjoSUJt7ENU0IkIEDCYD27G5goO_zfQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI14oL9i2wAQa7K6yEREmecBQB_z40_9_5z5qamIYTi9BA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 38912 + LdsInitCVgprs: false + LdsNumBytes: 38912 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38912 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MIVqWf-zHb4fGrqaI_PNfaMIuBSbf24WaOtDVKEmRwTy0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 166912 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 166912 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 8] + MIWaveTileA: 7 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 256 + MacroTileA: 224 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 224 + NumGlobalWriteVectorsPerThread: 224 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 8 + ThreadTileA: 28 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI2FE1_xdi6gisBeCfrvgksfzzbx9Yx6-LxacJqFoorOs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 71680 + LdsInitCVgprs: false + LdsNumBytes: 71680 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 166912 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71680 + LdsOffsetMetadata_Blk: 166912 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 7] + MIWaveTileA: 7 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 224 + MacroTileA: 224 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 196 + NumGlobalWriteVectorsPerThread: 196 + NumLoadsA: 7 + NumLoadsB: 7 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 7 + ThreadTileA: 28 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MItgCKUscKBBiMtBOmdwUt6qAHeM_MR7YbKWEkh39p8UQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 129024 + LdsInitCVgprs: false + LdsNumBytes: 129024 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 6] + MIWaveTileA: 7 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 192 + MacroTileA: 224 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 168 + NumGlobalWriteVectorsPerThread: 168 + NumLoadsA: 7 + NumLoadsB: 6 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 6 + ThreadTileA: 28 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI_UOU6HZrGnyaniPidkPyA2vc3eMgOLn6XHhdMyr3gok= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 5] + MIWaveTileA: 7 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 160 + MacroTileA: 224 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 140 + NumGlobalWriteVectorsPerThread: 140 + NumLoadsA: 7 + NumLoadsB: 5 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 5 + ThreadTileA: 28 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIdhuZzZrmslv6fIdikHYnhn1dpXNXnkfYYXLJoFZzt6s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 128 + MacroTileA: 224 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 4 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1b5RoEaxogDkty9pH0a4h4iqEK74VHsxBERA-H-NeUNY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 3] + MIWaveTileA: 7 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 96 + MacroTileA: 224 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 7 + NumLoadsB: 3 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 3 + ThreadTileA: 28 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1g3lxPhwHBbhj-5_vMz-ezv_Gct2IZefQGw3UsfUuGuM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 7 + NumLoadsB: 2 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1pRHgvlAR7N2hth7GOj12UHkPW1O97Tnl14UAnpJSv9c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 7 + NumLoadsB: 1 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICt7le9q16ngZmPo_ZNYkoWOjOg2jVjonbKgLeTEzLxQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 8] + MIWaveTileA: 6 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 256 + MacroTileA: 192 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 8 + ThreadTileA: 24 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MIFCEugnbAE1o_KNOxQTjmgk1-riRp7-bqr1XYLHQkDJI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 129024 + LdsInitCVgprs: false + LdsNumBytes: 129024 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 7] + MIWaveTileA: 6 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 224 + MacroTileA: 192 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 168 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 6 + NumLoadsB: 7 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 7 + ThreadTileA: 24 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MIsXAzSnuik23-u_Tqj9IA9yIVlTTHvUnKVAh5tdPfE-0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 120832 + LdsInitCVgprs: false + LdsNumBytes: 120832 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 6] + MIWaveTileA: 6 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 192 + MacroTileA: 192 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 6 + NumLoadsB: 6 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 6 + ThreadTileA: 24 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIdawCsxOV8Q1yfZkDghTj5ROObRZEHHk5r8TLSJ8-4DI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 5] + MIWaveTileA: 6 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 160 + MacroTileA: 192 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 5 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 5 + ThreadTileA: 24 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI96PVzcjbGKsmGAPHOizb5QovZZPMnBXqYWXYWCMR6RE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1XF9bGkyS2smPrKpZKRvXqZE_IlG80gFUg26F5mTZLis= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1cyGRGs5fNOjzNULtz2NEdDFcpvgD-53w5wck9Mz0rEA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1dXOVrMlvBwO6y7ESILkUB5p9GtB-Km9ykpVzg9YqhsY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 60416 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 60416 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI_FMKfIaIlw04mnTidFqT-Fnj0gDbPBKMLBy9AxGoipc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 124928 + LdsInitCVgprs: false + LdsNumBytes: 124928 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 8] + MIWaveTileA: 5 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 256 + MacroTileA: 160 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 160 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 8 + ThreadTileA: 20 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MIgvcBHckHBXx7rYEXvd58PMUeCwZSGsDJtMgKSLGOngQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 7] + MIWaveTileA: 5 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 224 + MacroTileA: 160 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 140 + NumGlobalWriteVectorsPerThread: 140 + NumLoadsA: 5 + NumLoadsB: 7 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 7 + ThreadTileA: 20 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIiC0cBFJQlkymU09JdEq7dUq1wvcSX9wghKOSQxkOErI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 6] + MIWaveTileA: 5 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 192 + MacroTileA: 160 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 120 + NumLoadsA: 5 + NumLoadsB: 6 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 6 + ThreadTileA: 20 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MIwREMu9dXpHu0hNi1Te8Ryumj0HYpn1BFkusQoNfhAQ8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 5] + MIWaveTileA: 5 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 160 + MacroTileA: 160 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 100 + NumGlobalWriteVectorsPerThread: 100 + NumLoadsA: 5 + NumLoadsB: 5 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 5 + ThreadTileA: 20 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIcrglZoxFbHHjwhvMHFjwAoFdYaccIqWMvEsbqy0yalI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1BUsuevlFp3qE9qBkCpl8ZHtyZp2nTMvZX1nz1cQRT7M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1IVM_kq6fQOztyL87pw0KKbmpiglUkaPXmejIGutUrh4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1bgH1ukcWIV-QEjZUrEMgbOOF-CIy8Yu_mpQb3L9Zq3g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 58368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 58368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU8_SUM1_SUS256_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM8_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 51712 + LdsInitCVgprs: false + LdsNumBytes: 51712 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51712 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [32, 32, 64, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 64 + MatrixInstM: 32 + MatrixInstN: 32 + MatrixInstruction: [32, 32, 64, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NonTemporalC: 4 + NonTemporalD: 4 + NumElementsPerBatchStore: 16 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 8 + StaggerUMapping: 1 + StorePriorityOpt: true + StoreSwapAddr: false + StoreSyncOpt: 4 + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 2 + SubGroup1: 128 + SubGroupA: 2 + SubGroupB: 128 + ThreadTile: [1, 1] + ThreadTile0: 64 + ThreadTile1: 2 + ThreadTileA: 64 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 1 + enableLDSTrA: 0 + enableLDSTrB: 0 + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIHAxeqAl5GZyYpWyj9YBlHLiAot0B1CVx9Nf_5OHRC1Q= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 224 + MacroTileA: 128 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 7 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIIlrnP4n1WnSi3fe0xoxepRfTuuLMMp-jgbnYdGnnvYE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI5bZ7yznefmkxpzzbB-XiEKlKRJu2AMeEgvXx60Tq5L0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI72iEPgzvyM2ZrWa9mJE7BjRqJtpIrgM81I_tGNEjK0c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1xf_ukgQ-Z-qNS4jmkabhCAr3mB7WAQFBVo0jxB_s6wM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1j8uVA7i_pGx8ZCBrMR8ZhK3U6bp7J7YjtbYnnKdPxDI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1M6uR2wC2h6sr2Hl9YrF8zhWGQ7RgoVsQ-TM7lCE6pPI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 22528 + LdsInitCVgprs: false + LdsNumBytes: 22528 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22528 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1lw4esJJcKXjrN7F_gKTEASvr4FPioQKVPOZImphrvBY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 49152 + LdsInitCVgprs: false + LdsNumBytes: 49152 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 49152 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1JixbiZckG_HlXVPjjWZHNInPUwLf2xyuU7BTFa5yHI0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 7] + MIWaveTileA: 3 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 224 + MacroTileA: 96 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 3 + NumLoadsB: 7 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 7 + ThreadTileA: 12 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1wIqbeUYowBvH47eY0D90voHkVaoBg2lPdbX2d68nIJ4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1J-2M-3q4LSCntZjUh3v8nZH3qUcDC89rCJ0Fdq3iMeI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1Vr1CG6o8iRf8BbhZJxSu0FVBXoX31LtlPuM8cs_YEXU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16Cnahg_B3KmoGxZkTO1OYZLBTHOrwHNY1VQd2t5IrylI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16W4yMmcRT_Rcp3ZNo2gSW3q4SJdd9i9DRRKcgqox59VE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 24576 + LdsInitCVgprs: false + LdsNumBytes: 24576 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 24576 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16zNms4_iu6xgFQiq42WUbXrNnFNyslp04sbAKWAlTbog= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 20480 + LdsInitCVgprs: false + LdsNumBytes: 20480 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20480 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI11aFcbk23grO4-hK8aBkA0Q2TJ_RzSnSefhCsshDFWE8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1BIQ9uHrC9E8FHJ9mzReo0log_q-mFe47Gvfw9lRTxKU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 7 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1oWIDEj_bU2sIdW6WZSqKJI8Okx_mH0MwWvgKg4p0Wh0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1DfS6ucItLGFYy3q0Zz-JZd7icG_ycHgazunawsT8Pns= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1jK-MUAAqzxLFJYDxWtp_lRnOH6G_Sa3LS7gsDkzJ5ZI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16qBSTIiXAL1b2k0UyETAi0pu4GSfWwKXX5o-ybzBtncE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 24576 + LdsInitCVgprs: false + LdsNumBytes: 24576 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 24576 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16mOGETtdY_W1srxs4Mfgd04W-cV79aCR6raMD3HA6jWU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 18432 + LdsInitCVgprs: false + LdsNumBytes: 18432 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 18432 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16wIclDOa00a_BcP3sOUDpDc_qFY9RvbDE777DV1a9O7k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 25600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 9216 + LdsOffsetMetadata_Blk: 25600 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1pg1aJvOuqnmTxe_-2e9_1S4uFGoAaPhgR-zs28ui024= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 38912 + LdsInitCVgprs: false + LdsNumBytes: 38912 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38912 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1bJJ5iziycwP81Vg_rFbnRobq-WE2Gw-XuD7kvGV58LU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 1 + NumLoadsB: 7 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1c60kFRwy-kwCbpVB3Be-Tp-X0igfgGU9wMtdCOAL5xM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI13_eugo-sWMKarMTBe1WI4hmzJdiGvToykcHBbj4WXow= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1gSIhM4a60TQQmpGca3zPH_1C9YMOrBSb2Bu9JrxvD-k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 22528 + LdsInitCVgprs: false + LdsNumBytes: 22528 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22528 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16YH7g6N9l5w5imlP3Lp9mdDXwtO_ZtzNO9yUXMFVOjKQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 20480 + LdsInitCVgprs: false + LdsNumBytes: 20480 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20480 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16IFZMDGDs3-q6koIU2xIB_XUAfzhmng9Mg_46GO3Rqf4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16rLKaVqye5ZVlglTUO9ioHrZ6WB-GJ_Qs2RqP4S8ZIXE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MIyoGz0G9DWmdNz6WjS2Bq6rCsCrR_mXM4td8eMkMYEDw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 73216 + LdsInitCVgprs: false + LdsNumBytes: 73216 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 38400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 165888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73216 + LdsOffsetMetadata_Blk: 165888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 15] + MIWaveTileA: 4 + MIWaveTileB: 15 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 240 + MacroTileA: 256 + MacroTileB: 240 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 240 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 8 + NumLoadsB: 15 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 15 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 15 + ThreadTileA: 16 + ThreadTileB: 15 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIBnc7kWhy6AvjwbNc5adjziCfJiLAVAD1RzoNcS6OMZQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 68096 + LdsInitCVgprs: false + LdsNumBytes: 68096 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 165888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 68096 + LdsOffsetMetadata_Blk: 165888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 13] + MIWaveTileA: 4 + MIWaveTileB: 13 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 208 + MacroTileA: 256 + MacroTileB: 208 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 52 + NumLoadsA: 8 + NumLoadsB: 13 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 13 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 13 + ThreadTileA: 16 + ThreadTileB: 13 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIF5A-1D1ZIPjJsJJ7ttt4AlfEFPQG0riur7Zv_qXG5YI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 128512 + LdsInitCVgprs: false + LdsNumBytes: 128512 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 28160 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 11] + MIWaveTileA: 4 + MIWaveTileB: 11 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 176 + MacroTileA: 256 + MacroTileB: 176 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 44 + NumLoadsA: 8 + NumLoadsB: 11 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 11 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 11 + ThreadTileA: 16 + ThreadTileB: 11 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIyRCs1CYgk6qoLPVzZz2698DWR4wMJ7whNVJ8JwJfnZc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 123392 + LdsInitCVgprs: false + LdsNumBytes: 123392 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 9] + MIWaveTileA: 4 + MIWaveTileB: 9 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 144 + MacroTileA: 256 + MacroTileB: 144 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 8 + NumLoadsB: 9 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 9 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 9 + ThreadTileA: 16 + ThreadTileB: 9 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIaVDafPaNDXaLtu56b1YkZsQUI1aLmxb3jgdu_dv65EA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 17920 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 112 + MacroTileA: 256 + MacroTileB: 112 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1jNVYmoC6fjSjbIADTf0f9l0Cs4TKifqjd6Xql0ehllk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 47616 + LdsInitCVgprs: false + LdsNumBytes: 47616 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 12800 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 47616 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1eX_SljyDEcS5rCPMhk2caJwb3hhe7qqzmb_HRJ0i-U0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 7680 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1fPZ5I9MI2WZKscvBb8L9C1UGdH6oze92UNxsd_rmjZ0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 37376 + LdsInitCVgprs: false + LdsNumBytes: 37376 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 2560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37376 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI17whl_3k4Y1EhiqZzn7y-o-UTqlat-Sd3AsiBr6V3nc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 73216 + LdsInitCVgprs: false + LdsNumBytes: 73216 + LdsNumElementsAlignedA: 38400 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 38400 + LdsOffsetB_Blk: 169472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73216 + LdsOffsetMetadata_Blk: 169472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [15, 4] + MIWaveTileA: 15 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 240 + MacroTile1: 256 + MacroTileA: 240 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 240 + NumGlobalWriteVectorsPerThread: 240 + NumLoadsA: 15 + NumLoadsB: 8 + NumLoadsPerpendicularA: 15 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 60 + ThreadTile1: 4 + ThreadTileA: 60 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIjK835T0h-3Ar1N2b4kmmt65KtYLTJhfaZH3LmR2auFA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 68096 + LdsInitCVgprs: false + LdsNumBytes: 68096 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 68096 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [13, 4] + MIWaveTileA: 13 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 208 + MacroTile1: 256 + MacroTileA: 208 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 208 + NumLoadsA: 13 + NumLoadsB: 8 + NumLoadsPerpendicularA: 13 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 52 + ThreadTile1: 4 + ThreadTileA: 52 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIgsJynaoGp5Lr3OC9AUIiNlgAW9kkq7qPUmgK4ohOA_k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 128512 + LdsInitCVgprs: false + LdsNumBytes: 128512 + LdsNumElementsAlignedA: 28160 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 28160 + LdsOffsetB_Blk: 93696 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 28160 + LdsOffsetMetadata_Blk: 93696 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [11, 4] + MIWaveTileA: 11 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 176 + MacroTile1: 256 + MacroTileA: 176 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 176 + NumLoadsA: 11 + NumLoadsB: 8 + NumLoadsPerpendicularA: 11 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 44 + ThreadTile1: 4 + ThreadTileA: 44 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI6ZUiH-qsm4LTRuwyb7uw6oTXUaPI3Vjpt7WHaUbBOfI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 123392 + LdsInitCVgprs: false + LdsNumBytes: 123392 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 88576 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23040 + LdsOffsetMetadata_Blk: 88576 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [9, 4] + MIWaveTileA: 9 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 144 + MacroTile1: 256 + MacroTileA: 144 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 144 + NumLoadsA: 9 + NumLoadsB: 8 + NumLoadsPerpendicularA: 9 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 36 + ThreadTile1: 4 + ThreadTileA: 36 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIiu6Dap7c-mOezayg2ePWEWJWIUNWwq_GUAWuc6asW8k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 17920 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17920 + LdsOffsetB_Blk: 83456 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 83456 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 112 + MacroTile1: 256 + MacroTileA: 112 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1fyJI8fnAEv3sOPHcVPqOC2AU9kSqOftCPxastr6HZJE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 47616 + LdsInitCVgprs: false + LdsNumBytes: 47616 + LdsNumElementsAlignedA: 12800 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12800 + LdsOffsetB_Blk: 78336 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 47616 + LdsOffsetMetadata_Blk: 78336 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-Zo3r6x3T98ZMWTZ4H7LqLtk7xXmH_6zOjhPN09FdTk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 7680 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 7680 + LdsOffsetB_Blk: 73216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 73216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1wehOWUghk2A68IMMHO37hmuhc-kFrA6X9yfb9WRqhBg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37376 + LdsInitCVgprs: false + LdsNumBytes: 37376 + LdsNumElementsAlignedA: 2560 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2560 + LdsOffsetB_Blk: 68096 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37376 + LdsOffsetMetadata_Blk: 68096 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI1vCzuwEprjPexlyglCW3zRXsjX9xUmGeZzokhH60Z9yE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 75776 + LdsInitCVgprs: false + LdsNumBytes: 75776 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 75776 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 2 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI1SiLGOArtDjYGdvuIvIxglRBR2PASF7znHipyq968J9k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 81920 + LdsInitCVgprs: false + LdsNumBytes: 81920 + LdsNumElementsAlignedA: 64512 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 64512 + LdsOffsetB_Blk: 195584 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81920 + LdsOffsetMetadata_Blk: 195584 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 14 + NumLoadsB: 4 + NumLoadsPerpendicularA: 14 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI1Atkrqghs5hZ7rFrRDMn1qRHk9P7RjRIX2jTHJaCZcIs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 64512 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 64512 + LdsOffsetB_Blk: 195584 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 195584 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 14 + NumLoadsB: 2 + NumLoadsPerpendicularA: 14 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI1H6kBMzYhDsiXY1vBEU2Ud8UhLYehCYuWeQaCulrUTHo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 12 + NumLoadsB: 6 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI153vNbJPjGrVniCPDOzS5UGzSK6NOLhTT3dYrLoo1YJ4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 12 + NumLoadsB: 4 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1y0BaulKQmTtKr2XSp5P3dnaOK0k5UYUxfkhT1HxC_rU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 117760 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 117760 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 2 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MIUYoy0JGLYkrhIavz9gjog-Elk__Z0jYq-QXQEcN2PaA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 177152 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 177152 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 10 + NumLoadsB: 8 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI1cUVPZBoZ-mtmE8sTWF_Az_FU-gB_6I8zhwfiN1pEaks= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 177152 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 177152 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 10 + NumLoadsB: 6 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 88 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI1EM3GcfaRjwnLLnXwjtnf8XZHB699buVut-TXk4ZjWyc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 63488 + LdsInitCVgprs: false + LdsNumBytes: 63488 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 63488 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 10 + NumLoadsB: 4 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 89 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI15N7cdBjXALKtpaYkT17ouFsMqBdqYm1rqUc2cC7GeLE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 10 + NumLoadsB: 2 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 90 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MIU14cbwZ_t9NC-mBYr7N5MljJmeoGwBb-WUmjyy7ILhU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 10 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 91 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIM08mmLtqT14x3TxGefabTcO290EciaeH6GnjDeU8Bgo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 92 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1OG3k0SIeaxxBYmDDXCQJ0v824MXDB2cvmAH14Io8mlM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 93 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1Ee_Jq2fBrVEHXaPp-qikzlaFXpjSm2wKDPvD-MKxLM0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 94 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI13IeiqjCNSTMel5JBLjnHPiB9PE_JYWIIfXqePFminMc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 95 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI1qUn-crPudL6dXZ6rCcKnkE9s4w-RcuNNqOhYdbIrR28= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 158720 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 158720 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 6 + NumLoadsB: 12 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 96 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI18wp9zZc9oencMlWPaDY7BUc_iQgMxwF-RQOzHet7J3I= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 158720 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 158720 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 10 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 97 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI1LrtbILHycv46qRt6NxJYDFm4zT_v-o9V1TAKrHNeAeQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 98 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI164Fx4NhdfXwLWoN6X-Afen5a5uy0a8LeKZgpjvani3CM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 6 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 99 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16_Nc2lY4fwUC-cbMivJ1U_eNf_9vhMW0WYu6jNMKBzYI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 100 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16X2E4Wx9NVLCyWp4uy3bGT7QFxR20eVc2o2HQJqwi8Hk= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 101 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI1ehl3l0UKCFUW0CqlYWmD5Gv6DpebCzV_7KPnbR_5y4Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 81920 + LdsInitCVgprs: false + LdsNumBytes: 81920 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 64512 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81920 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 14 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 14 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 102 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI1PpwSxkO--tNKZnLh5u1Zqztz5GRs2qwDUtmWoQ32_0M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 12 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 103 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1BIcxM2OYKG0Jqn5GRpKyHvCNhj2aQceoHMvTm7lJo6Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 63488 + LdsInitCVgprs: false + LdsNumBytes: 63488 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 63488 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 10 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 104 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1-eL7HedP-uoMH-4uLF-vatbGDwtmp3GZhw8IuLzDHvE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 105 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16Tidk1Ou4UA72T_vnHnWAsymx18z-7WG5S7yV6EMzxAE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 106 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16P7nrnDgCqsPXJVFvUTHGSoy-i5STapWZDg4F8KHN2a8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 107 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI164FNCErQfeGGweRYwOlRLTlQQw-uihsMini7VG5MjqUs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 108 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI1Weg33rA_lR82ctIL5Wv-agG74LvDUGtagD8ZNXCm6m8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 75776 + LdsInitCVgprs: false + LdsNumBytes: 75776 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 140288 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 75776 + LdsOffsetMetadata_Blk: 140288 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 16 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 109 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI1yjWOfHPT7EqX9rhhXejY8_eiJ07brJZxCVBlCXR9DBo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 64512 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 140288 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 140288 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 14 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 14 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 110 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1pVP6pjfyPWPVeerMg8FY1EyiGjH4B_j7tc4lRUzRPvg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 12 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 111 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI1q7mTED5UvVJrOPDvYFNPflDjUkX1okY5IJT3AZBddMM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 10 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 112 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1fFpYWYjPWsmrBYgoz9CIhz7OHVoysPvt9SgTel6GPWI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 113 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16mu07e3PYZ_jp2LQxX0lmvu4pmsw_265fxGpIWycVSjg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 114 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI162x9lNzvJGF2jhOPNNRp1Q3mFaUkUkcNWlM7ecLmXXfs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 115 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16-01AyvDsXfpTy-wvoK3K25-ZRKuFeCLIRsj5lVXzX_s= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 18432 + LdsInitCVgprs: false + LdsNumBytes: 18432 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 18432 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 116 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI1LDs52gy7NI4WuOMQ_idbNXBwBK86orsq1YVyOo4IvKs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 90624 + LdsInitCVgprs: false + LdsNumBytes: 90624 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 90624 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 16 + NumLoadsB: 5 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 117 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1NztuuZzeaShCMYTqz64dRhvSvIEetpxjBMHS12EujLs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 81408 + LdsInitCVgprs: false + LdsNumBytes: 81408 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 13824 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81408 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 16 + NumLoadsB: 3 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 118 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1-eS3GREuZbaF_WcdTrJ8tJN1VRBcjitHDgD1jEhWCcE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 72192 + LdsInitCVgprs: false + LdsNumBytes: 72192 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 4608 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 72192 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 119 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI1ZLUtUYYiOHV_3Bx1CweJ56xkOj29nbS2TUKPiIsgWeM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 90624 + LdsInitCVgprs: false + LdsNumBytes: 90624 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 154112 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 90624 + LdsOffsetMetadata_Blk: 154112 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 16 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 120 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1QlOH2GDodFrn6A0rxCKHrFBaU_QWfv580nm9xWdZY0k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 81408 + LdsInitCVgprs: false + LdsNumBytes: 81408 + LdsNumElementsAlignedA: 13824 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 13824 + LdsOffsetB_Blk: 144896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81408 + LdsOffsetMetadata_Blk: 144896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 16 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 121 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1PZ1TyY_JO23lBPKX9VhACTOX37NNgYiBjf2A-tanW7w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 72192 + LdsInitCVgprs: false + LdsNumBytes: 72192 + LdsNumElementsAlignedA: 4608 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4608 + LdsOffsetB_Blk: 135680 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 72192 + LdsOffsetMetadata_Blk: 135680 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 122 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16Z6LYIKVcBKQJUXMfCTkH2UmyhUWm90vrVgFH5F5h_is= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 4 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 123 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16Gx0c38MGhJUAjADEf6sgUUCYEnSxT0HvEevn-TLzDBA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 124 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 125 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16tO84-u8CmDozhzOYFPjZx1aAnLl_RdoGaIlGdosGUzM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 12 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 126 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 127 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16nxXVLq-_LhX2N8-ggvlvGFb2CSQPdL0Jk31-3Yn8K8w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 128 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI1Uh3YqlpEBhxKyjcdbTJSpHtEarJLrJfEIa0Mnxwsb5k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 8 + LoopUnroll: 1024 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 129 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16t0-mPljiEZtupRhWAFpnxBDrJen-FsvDD34Q7PN3s_4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 130 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16CAJPr2P6olRKJqA2Fmiymyoh6JhX7nYYGS7ffsGJvmw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 131 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16K_Na50sAdM91rhXevAlyiM9lrmobRF_PSYV6N76rQzk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 132 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16dVDH5F1pzH8NV-uymOAiVxprnNui5yf2Mp5t1_LqLTs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 133 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16Uv-d7icfFxrZ-YUiRaTBsX3xXbc3HcmSMlEdtwh8Yb0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 134 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1TNTLRLqQ0uhz1zq5XPt3rxttGE1MT5QoFsMWFmpwZGg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 135 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI13HHr0kFgQAqJ2TjtZ8tTohe6Udz-5u0CwkMq-_9renQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 136 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1KAql7WR7THsIAKFHzmlcHQVpFofe0Vzt2xFE0wiTOfU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84480 + LdsInitCVgprs: false + LdsNumBytes: 84480 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84480 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 4 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 137 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1-LGVgPAQHxTiHAOxIZA3A-LLqgvwFhsRG4LoerQvYqg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 138 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1uwD3q8H7ROxJTYYpP8cvtJiz3DeoGvbO5IiJaZpCFnE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84480 + LdsInitCVgprs: false + LdsNumBytes: 84480 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 147968 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84480 + LdsOffsetMetadata_Blk: 147968 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalA: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 16 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index eaa249906182..227f8102c782 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: true +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -79,11034 +81,8043 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1zEvSrUfniyIPA5Nficsug_rlEaVAGxHNYFMuvmYifqc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1oq7eVBWwzey19A8uIqsDkLEvAC8u4jm_3WdeHmyqNY8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1RipBYLITmgLj7nJiXhyI57h-QdswAlXmvZRQ-t7Yij4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33280 - LdsInitCVgprs: false - LdsNumBytes: 33280 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1iIhyCvIYW4EQ_x86i0IZ86zDKGdESkaJ9YYcm0hSGzI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 57600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 57600 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1eZeOxd9MrHLQR2GNA65EfarGb_PDYi0bDvnnTj3iymw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1fok8NkYtMokxqZZMiJ_iGKUDQvUnmdxY1flvKGmyHyE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI4THos_lFJzWi_Dnch9qAlPGNCCGbpGTJT_pt2NL4ZB8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1pEJy9yEKZBXA9EcPO5rjF-8vFHW5-X8UkNyH2oVn994= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29440 - LdsInitCVgprs: false - LdsNumBytes: 29440 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29440 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1S6e2ViSr3lOZh36puDEQ4ogQb2Kj5RPE7WzL1lq3F0M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1C6jTq6qNnho1ENixE4x2q_23xMvUAX0Q3bKaL_ah5XQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21248 - LdsInitCVgprs: false - LdsNumBytes: 21248 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21248 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1158FxB1URc5cpKzdBy5PKmwBlaNnYvU5pNK2aJTi3FY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29440 - LdsInitCVgprs: false - LdsNumBytes: 29440 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29440 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI163lHbClsVTGPnNSzP8QvKwmggFBp2OvBLf9IoiXlJZAY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16cyaq9MxU7sNbLEazHdXIfoficQy4YOPE4TFqMztilgY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20992 - LdsInitCVgprs: false - LdsNumBytes: 20992 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20992 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16O3R5PeDu0JjUKDdGyCc1CoE20Q-U4WqcAOA_B9UwqIo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1LPyOUj4mpprhT6Es7dZqjCKeRoX1IDrnjJ3fYZl1HPI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1dsiiVyF-YL7QpBTxukjNXBE5_N2gPbhGhfp7ioSoYBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33280 - LdsInitCVgprs: false - LdsNumBytes: 33280 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1w-MN2Mm1J0St8q8s9xFauQAWxA0khy1Pd4WoSoekSlg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1cgKWeOcB-B078ImMnsIo0ZS6zy_3JERNpCWjdkytEyw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16_rKHyj2ip3IiAlJ-oI4Mi3S3Jk5OCWj2H83MWBg-Vss= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20992 - LdsInitCVgprs: false - LdsNumBytes: 20992 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20992 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16Qa_B98mEcapAL2iX2YgOcoJisssJyHDlM9u9CPAkaUM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16EwXWjXKR9ATyJNfH3DyoYi3LQ4H0ugLmKOZi9zzkKhc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29056 - LdsInitCVgprs: false - LdsNumBytes: 29056 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 24832 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 8448 - LdsOffsetMetadata_Blk: 24832 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI15LrCqHq2DyagcJd2X5qTNiDHWQnVHm_AqMqcFDlvmsU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1RS5-E3dcQYltJiaIWglbCOB49DJT8EKGQr29y9ht3jk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1WAM7dQoYboHh0SapFH0XK75zgHRfXzHLhlFGHRFgW8U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1fkykvXwBDcTCeZQcsORsfStQSOevaJqa0Sia79tVqpE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21248 - LdsInitCVgprs: false - LdsNumBytes: 21248 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21248 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16WEwxgg8Y4FSUK3HPuX6yKvRQPHQ0p3bui3ZyCjYgs5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16PYalQdoVElqz6wbixR1EPrpAfE0UNhfwtrN4ZWmHKYs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16XYtEgd7zoKly5GXv2t2pAYLX-e76OQRyOB6Ygozeu0A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24960 - LdsInitCVgprs: false - LdsNumBytes: 24960 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1JWVR-jh-aLg0rrdtghL_Hme5Pul5fnhSrwkCOBuPtIU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 768 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 39680 - LdsInitCVgprs: false - LdsNumBytes: 39680 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 6400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 39680 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1PwJ42uBJtJDXBA5J9eUpFXnSBdG873yhqH9x9xHr9aM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35584 - LdsInitCVgprs: false - LdsNumBytes: 35584 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 2304 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35584 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-ckiACWa5_Hl9q0vruUn7fAZYeT9xW9mnDjkhP2c50s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 768 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 39680 - LdsInitCVgprs: false - LdsNumBytes: 39680 - LdsNumElementsAlignedA: 6400 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 6400 - LdsOffsetB_Blk: 71936 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 39680 - LdsOffsetMetadata_Blk: 71936 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1MnL3tvF_jGDAY_I8anP7ysG0uEcoQof1urIS_BrpX_k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35584 - LdsInitCVgprs: false - LdsNumBytes: 35584 - LdsNumElementsAlignedA: 2304 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2304 - LdsOffsetB_Blk: 67840 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35584 - LdsOffsetMetadata_Blk: 67840 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1arXBxIFzyAilDtP1DqyksxdPITOkluw9jmLU5u0Tt2w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1rXabrsjNwkup_JrxozMVR_3U4GdmTFKff0KArypEYAA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1VT_L439A1JYsr4mdupck3knPO5y9YC1PTbi05SL_Cz0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16zQIQS-CrV5B-hJqIIcfIrW4-99zhrrQJDstr_sFS9t4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16yzgS8-MSN2UCxzxDcSy8jzQVwvB-olaZ82o1J7M6llU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI12hFMSt6lMTI8oWW_7aZTwxjiukY6LB_x-MEbbxd5p3M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI163QE_QBcfCjhjm6YvZW1aBPJMqF6E2nryKoAjsugzB0U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16CqP0rVSFO3WDeB13qT6LUGERdrWprMFCekioRFqiTqQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1h8DSl_KtVz4vyL0LXehy8GQa575LmnT-GS8Sm349l8c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 256 - LVCA: 16 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 70912 - LdsInitCVgprs: false - LdsNumBytes: 70912 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 70912 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1ABd-H42vx6NZOkifL5ShxVhgON-oxRsJ0GyaVxfUhIg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 256 - LSPB: 16 - LVCA: 1 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 70912 - LdsInitCVgprs: false - LdsNumBytes: 70912 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 135424 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 70912 - LdsOffsetMetadata_Blk: 135424 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI160Nuqi97WrEd5kWYgnYYkOYOWZ8I-jc1R0TyAYU3gxQk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 16 - LSPA: 256 - LSPB: 256 - LVCA: 1 - LVCB: 1 - LVPA: 16 - LVPB: 16 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16y2XC2HjmywvRNMRzvdc4xwYo7V4GZ1mS8mdECpGkL28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 16 - LSPA: 128 - LSPB: 256 - LVCA: 2 - LVCB: 1 - LVPA: 8 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI162YLFRDLjA2hwymDeM7kI4wEg1_Ytlga74L2JT4VF0MM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 64 - LSCB: 16 - LSPA: 64 - LSPB: 256 - LVCA: 4 - LVCB: 1 - LVPA: 4 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI1651AQfSOaqL9x7qKXZbPBT7ZWUW0NRJycR149B4OSQTc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 32 - LSPA: 256 - LSPB: 128 - LVCA: 1 - LVCB: 2 - LVPA: 16 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI1674YMvMGkXuXL1Cf1gzLE5KEGPC_Zl12HLRkso0LE7_M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 64 - LSPA: 256 - LSPB: 64 - LVCA: 1 - LVCB: 4 - LVPA: 16 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1vOk7fZIE6d2Crn4j8R3f-gr9gYb6d1i7QRitxgT49e8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 16 - LSPA: 256 - LSPB: 256 - LVCA: 1 - LVCB: 1 - LVPA: 16 - LVPB: 16 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xLaYt4XA53nJMoN4GRDxGCoYkMto7pvV9s9Hmk9M424A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1zEvSrUfniyIPA5Nficsug_rlEaVAGxHNYFMuvmYifqc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1oq7eVBWwzey19A8uIqsDkLEvAC8u4jm_3WdeHmyqNY8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1RipBYLITmgLj7nJiXhyI57h-QdswAlXmvZRQ-t7Yij4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33280 + LdsInitCVgprs: false + LdsNumBytes: 33280 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1iIhyCvIYW4EQ_x86i0IZ86zDKGdESkaJ9YYcm0hSGzI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 57600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 57600 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1eZeOxd9MrHLQR2GNA65EfarGb_PDYi0bDvnnTj3iymw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1fok8NkYtMokxqZZMiJ_iGKUDQvUnmdxY1flvKGmyHyE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI4THos_lFJzWi_Dnch9qAlPGNCCGbpGTJT_pt2NL4ZB8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1pEJy9yEKZBXA9EcPO5rjF-8vFHW5-X8UkNyH2oVn994= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 29440 + LdsInitCVgprs: false + LdsNumBytes: 29440 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29440 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1S6e2ViSr3lOZh36puDEQ4ogQb2Kj5RPE7WzL1lq3F0M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1C6jTq6qNnho1ENixE4x2q_23xMvUAX0Q3bKaL_ah5XQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21248 + LdsInitCVgprs: false + LdsNumBytes: 21248 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21248 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1158FxB1URc5cpKzdBy5PKmwBlaNnYvU5pNK2aJTi3FY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 29440 + LdsInitCVgprs: false + LdsNumBytes: 29440 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29440 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI163lHbClsVTGPnNSzP8QvKwmggFBp2OvBLf9IoiXlJZAY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16cyaq9MxU7sNbLEazHdXIfoficQy4YOPE4TFqMztilgY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 20992 + LdsInitCVgprs: false + LdsNumBytes: 20992 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20992 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16O3R5PeDu0JjUKDdGyCc1CoE20Q-U4WqcAOA_B9UwqIo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1LPyOUj4mpprhT6Es7dZqjCKeRoX1IDrnjJ3fYZl1HPI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1dsiiVyF-YL7QpBTxukjNXBE5_N2gPbhGhfp7ioSoYBo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 33280 + LdsInitCVgprs: false + LdsNumBytes: 33280 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1w-MN2Mm1J0St8q8s9xFauQAWxA0khy1Pd4WoSoekSlg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1cgKWeOcB-B078ImMnsIo0ZS6zy_3JERNpCWjdkytEyw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16_rKHyj2ip3IiAlJ-oI4Mi3S3Jk5OCWj2H83MWBg-Vss= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 20992 + LdsInitCVgprs: false + LdsNumBytes: 20992 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20992 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16Qa_B98mEcapAL2iX2YgOcoJisssJyHDlM9u9CPAkaUM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16EwXWjXKR9ATyJNfH3DyoYi3LQ4H0ugLmKOZi9zzkKhc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29056 + LdsInitCVgprs: false + LdsNumBytes: 29056 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 24832 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 8448 + LdsOffsetMetadata_Blk: 24832 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI15LrCqHq2DyagcJd2X5qTNiDHWQnVHm_AqMqcFDlvmsU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1RS5-E3dcQYltJiaIWglbCOB49DJT8EKGQr29y9ht3jk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1WAM7dQoYboHh0SapFH0XK75zgHRfXzHLhlFGHRFgW8U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1fkykvXwBDcTCeZQcsORsfStQSOevaJqa0Sia79tVqpE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 21248 + LdsInitCVgprs: false + LdsNumBytes: 21248 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21248 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16WEwxgg8Y4FSUK3HPuX6yKvRQPHQ0p3bui3ZyCjYgs5k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16PYalQdoVElqz6wbixR1EPrpAfE0UNhfwtrN4ZWmHKYs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16XYtEgd7zoKly5GXv2t2pAYLX-e76OQRyOB6Ygozeu0A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 24960 + LdsInitCVgprs: false + LdsNumBytes: 24960 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1JWVR-jh-aLg0rrdtghL_Hme5Pul5fnhSrwkCOBuPtIU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 768 + LdsBytesNoAmax: 39680 + LdsInitCVgprs: false + LdsNumBytes: 39680 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 6400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 39680 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1PwJ42uBJtJDXBA5J9eUpFXnSBdG873yhqH9x9xHr9aM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 35584 + LdsInitCVgprs: false + LdsNumBytes: 35584 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 2304 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35584 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-ckiACWa5_Hl9q0vruUn7fAZYeT9xW9mnDjkhP2c50s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 768 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 39680 + LdsInitCVgprs: false + LdsNumBytes: 39680 + LdsNumElementsAlignedA: 6400 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 6400 + LdsOffsetB_Blk: 71936 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 39680 + LdsOffsetMetadata_Blk: 71936 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1MnL3tvF_jGDAY_I8anP7ysG0uEcoQof1urIS_BrpX_k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 35584 + LdsInitCVgprs: false + LdsNumBytes: 35584 + LdsNumElementsAlignedA: 2304 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2304 + LdsOffsetB_Blk: 67840 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35584 + LdsOffsetMetadata_Blk: 67840 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1arXBxIFzyAilDtP1DqyksxdPITOkluw9jmLU5u0Tt2w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1rXabrsjNwkup_JrxozMVR_3U4GdmTFKff0KArypEYAA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1VT_L439A1JYsr4mdupck3knPO5y9YC1PTbi05SL_Cz0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16zQIQS-CrV5B-hJqIIcfIrW4-99zhrrQJDstr_sFS9t4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16yzgS8-MSN2UCxzxDcSy8jzQVwvB-olaZ82o1J7M6llU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI12hFMSt6lMTI8oWW_7aZTwxjiukY6LB_x-MEbbxd5p3M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI163QE_QBcfCjhjm6YvZW1aBPJMqF6E2nryKoAjsugzB0U= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16CqP0rVSFO3WDeB13qT6LUGERdrWprMFCekioRFqiTqQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1h8DSl_KtVz4vyL0LXehy8GQa575LmnT-GS8Sm349l8c= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 256 + LVCA: 16 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 70912 + LdsInitCVgprs: false + LdsNumBytes: 70912 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 70912 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1ABd-H42vx6NZOkifL5ShxVhgON-oxRsJ0GyaVxfUhIg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 256 + LSPB: 16 + LVCA: 1 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 70912 + LdsInitCVgprs: false + LdsNumBytes: 70912 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 135424 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 70912 + LdsOffsetMetadata_Blk: 135424 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI160Nuqi97WrEd5kWYgnYYkOYOWZ8I-jc1R0TyAYU3gxQk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 16 + LSPA: 256 + LSPB: 256 + LVCA: 1 + LVCB: 1 + LVPA: 16 + LVPB: 16 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16y2XC2HjmywvRNMRzvdc4xwYo7V4GZ1mS8mdECpGkL28= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 16 + LSPA: 128 + LSPB: 256 + LVCA: 2 + LVCB: 1 + LVPA: 8 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI162YLFRDLjA2hwymDeM7kI4wEg1_Ytlga74L2JT4VF0MM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 16 + LSPA: 64 + LSPB: 256 + LVCA: 4 + LVCB: 1 + LVPA: 4 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI1651AQfSOaqL9x7qKXZbPBT7ZWUW0NRJycR149B4OSQTc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 32 + LSPA: 256 + LSPB: 128 + LVCA: 1 + LVCB: 2 + LVPA: 16 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI1674YMvMGkXuXL1Cf1gzLE5KEGPC_Zl12HLRkso0LE7_M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 64 + LSPA: 256 + LSPB: 64 + LVCA: 1 + LVCB: 4 + LVPA: 16 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1vOk7fZIE6d2Crn4j8R3f-gr9gYb6d1i7QRitxgT49e8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 16 + LSPA: 256 + LSPB: 256 + LVCA: 1 + LVCB: 1 + LVPA: 16 + LVPB: 16 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xLaYt4XA53nJMoN4GRDxGCoYkMto7pvV9s9Hmk9M424A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index 1b600d352af7..8992357845b6 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: true +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -79,19809 +81,14282 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIk_RqFM_AwoflFBuOWbvk-8fTqjDZ1FddXgJjbcy-G0k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 125952 - LdsInitCVgprs: false - LdsNumBytes: 125952 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 6] - MIWaveTileA: 8 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 192 - MacroTileA: 256 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 6 - ThreadTileA: 32 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MISPq2JndWaX4OL12RUN6ulsn7QcNRohtTIZC1Edgdyws= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123904 - LdsInitCVgprs: false - LdsNumBytes: 123904 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 5] - MIWaveTileA: 8 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 160 - MacroTileA: 256 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 5 - ThreadTileA: 32 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIKVXPdd7d_qWsOE3R4Nn3WRva2YN_nzMUEkSXMgI4hvI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50176 - LdsInitCVgprs: false - LdsNumBytes: 50176 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50176 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1T7zX-LAq1W4S0p0x8V1ahqv2mXqZFvICIzJD4wjn5vU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48128 - LdsInitCVgprs: false - LdsNumBytes: 48128 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48128 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1C8Q9_g8YRB2HVCaci6ZpxWtz2F-KOjbPdcOSgM2OeEI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1wcQqKyv34c9LbqUML817LBELS8jNDkpatWFuPGnAz0I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37888 - LdsInitCVgprs: false - LdsNumBytes: 37888 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37888 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIghxlKUHytHxV_rsO7Y3k-hKsNgsUv14Ji4M_f4zzLUg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50432 - LdsInitCVgprs: false - LdsNumBytes: 50432 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50432 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 5] - MIWaveTileA: 6 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 160 - MacroTileA: 192 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 5 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 5 - ThreadTileA: 24 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MIJBI5cmbVufum9IvszCzB765s6mkvxNBCOv5r57HrQ5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI195MLYeeCjfNjgEj-ik95KMPTRnsuVyYIPGD8iI7U3FA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1enbINkflafBFO3_Pebg7jg-XrgUNrjPKEtJ1FG_ArW4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1AXbZC25QBPC0PuokhYA5b8WyOR4HvuGr9lZsTR0HMWk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 57600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 57600 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIYn1IItDWJZY2hmD9-0eW5TADd3akO6yu4iNVSScCFLk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 86272 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 86272 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI15qZVVldYFEVxv6DoRCJq1o-8jbrdV19RqRcT-bcTp5g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 86272 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 86272 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1lHcr20MZlAXhwNOR6Gevj4wJaRc4B-tum0bYx49prt8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1XvrvtOqqq8hKEVOCgfEFmALGzVphawbwpkAxG7nqoi8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIYXJbKl66tz5Yz16uScZ-3QcnIi8M-PaVMoZpl5IQJR0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 8] - MIWaveTileA: 4 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 8 - ThreadTileA: 16 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIWMlYxh0gBBwfi1wvTDHJHkSEVXIuF12-m4Tq-srJGFk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 224 - MacroTileA: 128 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIU-MW4z63wAohwJUDz0C8OcgBozd0DUJEBLSSdararIA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIZLtCnUAcZQRb2Sm7BFv7A3a2vBOCNqxyJPub1dc8Z8U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MInMSU2Gh3mMVemcGZ3WqUWu2l4NG8N5rOCZZl_tZSxr4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1VCOEzWL6LPvmD0HkYK2IQmKeQ33MFQ6q3DD1gUWXPWI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32256 - LdsInitCVgprs: false - LdsNumBytes: 32256 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32256 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1bvrvP40X4hFkSo1KwPg0VEiYDXT6P8LD2PVpHtPr4t8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1KCNveOe_jYnymu00u65Iq4hQQmFEfEECKI4hYCSw86A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22016 - LdsInitCVgprs: false - LdsNumBytes: 22016 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22016 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI11tDVdUBSEo3UIS5TLyJqHVgQw3dto3agLBzaHthkf5A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46336 - LdsInitCVgprs: false - LdsNumBytes: 46336 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46336 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1uDCZdlP32w2aUIZQgU6UFaItvoNYlwEGmbTkXc1IEtc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48384 - LdsInitCVgprs: false - LdsNumBytes: 48384 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48384 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 7] - MIWaveTileA: 3 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 224 - MacroTileA: 96 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 3 - NumLoadsB: 7 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 7 - ThreadTileA: 12 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1tIdq_8HHc1AUHlfryN0_49Q8EAD9oEnK2lMgmhIKC2E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1o_9BqNra6zh_HmLvWFNQm4WYzN2W1bygUabZEe9Lzsg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1KePfYaceZ29tiGigvOtcCFJNsP7pueMsjzyHKVb7w64= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16tOU8-XLYrNPdvvuRaPZblzPTCq0IYpmOqRVYwuw2zIA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 27904 - LdsInitCVgprs: false - LdsNumBytes: 27904 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27904 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16iQaL89dfDqFaUwHSFuqBVL1d7gxxRqiMEISWe48rBhk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16G697WmrCkkuSW0ygicOEsjSabf2uABvGDaKBNngK4oE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1opMCZdZFzR31a1nXMExXx-HFGwt6FHBNysNBuXPo34o= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1GPhaxX2PStJ8QfX3S-AgQFNJD6o4d_c2qqsOdt_oaC4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44288 - LdsInitCVgprs: false - LdsNumBytes: 44288 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44288 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1gqbADNhoP3bHba8oMSfqNYB-76ixvzWU8E58C_oXxBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1-VJoMMTo08ubflQERsCoO0j_TzMOoFOBMG1VV9dqzpY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI15OARqHSqH8kbGBPru-B8g6Oad-Zo3FGsUS8SxkzZZYk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16xVRyCL8pcYWjhmBX3Sksf7_ECJZKzPUO0YM7WCFd9sM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 23808 - LdsInitCVgprs: false - LdsNumBytes: 23808 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23808 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI162emMCIrLY9uAeJtwZl6HlANVDofIksu3xpgoJdG4L7I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16UNManzuffvMTmsgfWnWF6Bt5jjvtS0tBsdToDQ_bZvc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 24832 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 8448 - LdsOffsetMetadata_Blk: 24832 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1kHn9o1wGnb0Le03rKuL4bApCYLkFlSLtoU6MjkRibzc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1NMWe8o4PG1BIiKP1ggEjZUpwkkJSDyvBBNMZ8tXiaR8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 1 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1Q0rm-ZJ-B1Ep3tDRV7Y6v5xFppIw1pc1wB6W7qCB6eE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32000 - LdsInitCVgprs: false - LdsNumBytes: 32000 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32000 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI14_XY01zUqMSgVV3Z29zc0HvqY2RBU1XaMixpmKPTFvQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1BV068w-wQamiyACMJrCQO6AzvkFtVVaL6YUOIybq_As= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI167ogjNu2arPxWNgqgKM8sDxj_QbeHe5G3biyDU4kITPE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 19712 - LdsInitCVgprs: false - LdsNumBytes: 19712 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 19712 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI1606uPSrg8jSF_adB8-Q1EUFxmHwf5F9lLMz64s3UmyCk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16Jm6coo6HlXoYc23v3SkS10L-uM9lwa5bnHnY2Vs_oLo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIv9D6qCD557UuGj0WYoB1KLme1Sh1mzoR8zvP-Ew1rDY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 66560 - LdsInitCVgprs: false - LdsNumBytes: 66560 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 66560 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 13] - MIWaveTileA: 4 - MIWaveTileB: 13 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 208 - MacroTileA: 256 - MacroTileB: 208 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 52 - NumLoadsA: 8 - NumLoadsB: 13 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 13 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 13 - ThreadTileA: 16 - ThreadTileB: 13 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIZTzgVMYbF8zAjkLbWPx_KguNjRPGDWzQphgfwu106mI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 28160 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 11] - MIWaveTileA: 4 - MIWaveTileB: 11 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 176 - MacroTileA: 256 - MacroTileB: 176 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 44 - NumLoadsA: 8 - NumLoadsB: 11 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 11 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 11 - ThreadTileA: 16 - ThreadTileB: 11 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIYpXgEPF2KmPhp_HSItkaLdu8ZfTql6wuzRMctpM8bWs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 121856 - LdsInitCVgprs: false - LdsNumBytes: 121856 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 9] - MIWaveTileA: 4 - MIWaveTileB: 9 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 144 - MacroTileA: 256 - MacroTileB: 144 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 8 - NumLoadsB: 9 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 9 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 9 - ThreadTileA: 16 - ThreadTileB: 9 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIM6ky9Tq6VRezat1v-Vy4Qy4OEEUGyFJHwE-XvlIpRaY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 17920 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 112 - MacroTileA: 256 - MacroTileB: 112 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1ZaT28UTzCJBM6wCUWArJk9cu4afCdA6oGhULOQC7jwQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46080 - LdsInitCVgprs: false - LdsNumBytes: 46080 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 12800 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46080 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1ckLt2U3Q4BSlwSpNR-cpd6uJ3sMr5OVPRWuKvEiSVtY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 7680 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1Cks6yoWu7mRHY5OU_L0wlyt_BkdBZ9B6KYLX0jMjEZM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35840 - LdsInitCVgprs: false - LdsNumBytes: 35840 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 2560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1LRieWm8obTv5BrugK6ZfaH9J4-zT3LrRBcx6GxwPdUk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 1280 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45312 - LdsInitCVgprs: false - LdsNumBytes: 45312 - LdsNumElementsAlignedA: 10496 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 10496 - LdsOffsetB_Blk: 76032 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45312 - LdsOffsetMetadata_Blk: 76032 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1zBVp9O_j-pDQmFJ0pruVZjPjyXjbaxIDNKuOKzE-hUk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 768 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 6400 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 6400 - LdsOffsetB_Blk: 71936 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 71936 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1LmEbjkFjRe2EHvxZ5njVJSBpnw78dqd37eC05zoq8oY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 2304 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2304 - LdsOffsetB_Blk: 67840 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 67840 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI_Cpsqq7IdCcI6m09ixnRrSz_HTRJbwY41LkpuZ6eaow= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1tP0yLw5sOaqSqLw69T0rUISjIec4gtRcEBnLindXxd0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1WlKSkQYWc45q9-9u3Sa8KR5G5YSwdMXXvYIooja5mzU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1ZFz2Wqi9AGPfO9YZeT8hu0U-a6QRpBHZ1MNZhOuWfzs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16YTIDOkvWpHTvJMbXMK3Qn2TUcHpLDIz3L67zYqwLtxM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1Hyiuad9E_LwogHYOQgHONBkR83eXTUE_-HwHw1z6c2M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 62976 - LdsInitCVgprs: false - LdsNumBytes: 62976 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 62976 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1_vkhGjNt0PNt_xV8VVOz2XzUCfwEcU8zljVlbKTICVg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16riuLixhIaRRQRAFH27-FM-5iTnYrSUKGh2HFdVNH4W8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16tRLsgniRf9NQ8cwUknJpPeoosk2I0euxrG_yXuRD6Zc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16zCjLhPMzebcvjlA5OUrujTBGN1OYulB3ktj1a570vLs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1OuevmgTEfRQM80_r7hzorlLPNIZiSHCEk0LhjgcOAH8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 60672 - LdsInitCVgprs: false - LdsNumBytes: 60672 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 60672 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI117HfbCeMXp_75IhCNgyVsAtudVa47Y080duX4tlb_Cc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 54528 - LdsInitCVgprs: false - LdsNumBytes: 54528 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 54528 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1w_TVfNrjE3mYvq96Bm1Jgkl-HCbXvzNQ6H6gr4PSnKc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16IX89BCLJtYzlNb-bQG51jUiesgW7gn9abBGKzaC4u7Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16yEC0PHvmD0zrKvRDTkhqsmiXq78VvD0Eg5fcAnocoic= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16_OtB8fTrZk3Pt0bE3bXfsWfkoWXEPa2ENPfe4eaiVRI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1uaMQi1863lUOUemwTE99mttXxpTRkvqvmiwvtfx6Qx0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 80384 - LdsInitCVgprs: false - LdsNumBytes: 80384 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 13824 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 80384 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 16 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1_aftRJpk6u1tTs3-Hwls-Fo3A3dD889Os-hWuvfNQvw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71168 - LdsInitCVgprs: false - LdsNumBytes: 71168 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 4608 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71168 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI18CjcSziK0PjBic2MAopWR-Ta_nHbDse6qNtds5pye7Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 256 - LSPB: 16 - LVCA: 1 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71936 - LdsInitCVgprs: false - LdsNumBytes: 71936 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 135424 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71936 - LdsOffsetMetadata_Blk: 135424 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 512 - LSPA: 64 - LSPB: 8 - LVCA: 4 - LVCB: 32 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI169OaDNKWZnXMa1R_AlPEgQpVqAFVb54TBTn3PvUMxY7A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 512 - LSPA: 128 - LSPB: 8 - LVCA: 2 - LVCB: 32 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16JRL6UviIw3ql2MWJlkjDhAPVUQuM-dfo_YP9V8INeCE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16a9rQBxUkyvDWlmJquyIGWqzMINaNpRD_aWL9QWIuzQ8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 512 - LSPA: 128 - LSPB: 8 - LVCA: 2 - LVCB: 32 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16Jyq1ax9kWbpfAk08jNUVf7z1XoXRVCVdpt-kBpa0fWY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 64 - LSCB: 512 - LSPA: 64 - LSPB: 8 - LVCA: 4 - LVCB: 32 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16D0scrs9_FOYkVHUHqfaWIBeu8Rz72d3pOV3R8iV8H74= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI169Ao_eCjaajChoPx9i0y3MGqMyeJdbqU3dbjwB_UIKx4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1DDaFTcewwV5Ybn8swfgQtfH13aPb98dLOnRrTIJYmSQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1ifWE51xli667eSWHdf8X0Gqm1isofKMCZj08QiVl3U0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 1024 - LSPA: 128 - LSPB: 4 - LVCA: 2 - LVCB: 64 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1U06udU6pkFuvB0exyiBHs-wOgMdXOgbcv4nENk4uw44= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1oDw0RAqdNEFhP7idqV0O87totk9ugWGoiXKFb7JRQVw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84992 - LdsInitCVgprs: false - LdsNumBytes: 84992 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84992 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xqwn4mYe9PqjtRIapRznDsflXKMnaPO3v98RcshyCRoc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 16 - LdsPadB: 8 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIk_RqFM_AwoflFBuOWbvk-8fTqjDZ1FddXgJjbcy-G0k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 125952 + LdsInitCVgprs: false + LdsNumBytes: 125952 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 6] + MIWaveTileA: 8 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 192 + MacroTileA: 256 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 6 + ThreadTileA: 32 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MISPq2JndWaX4OL12RUN6ulsn7QcNRohtTIZC1Edgdyws= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 123904 + LdsInitCVgprs: false + LdsNumBytes: 123904 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 5] + MIWaveTileA: 8 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 160 + MacroTileA: 256 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 5 + ThreadTileA: 32 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIKVXPdd7d_qWsOE3R4Nn3WRva2YN_nzMUEkSXMgI4hvI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 50176 + LdsInitCVgprs: false + LdsNumBytes: 50176 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50176 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1T7zX-LAq1W4S0p0x8V1ahqv2mXqZFvICIzJD4wjn5vU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 48128 + LdsInitCVgprs: false + LdsNumBytes: 48128 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48128 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1C8Q9_g8YRB2HVCaci6ZpxWtz2F-KOjbPdcOSgM2OeEI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1wcQqKyv34c9LbqUML817LBELS8jNDkpatWFuPGnAz0I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 37888 + LdsInitCVgprs: false + LdsNumBytes: 37888 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37888 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIghxlKUHytHxV_rsO7Y3k-hKsNgsUv14Ji4M_f4zzLUg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 50432 + LdsInitCVgprs: false + LdsNumBytes: 50432 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50432 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 5] + MIWaveTileA: 6 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 160 + MacroTileA: 192 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 5 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 5 + ThreadTileA: 24 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MIJBI5cmbVufum9IvszCzB765s6mkvxNBCOv5r57HrQ5k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI195MLYeeCjfNjgEj-ik95KMPTRnsuVyYIPGD8iI7U3FA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1enbINkflafBFO3_Pebg7jg-XrgUNrjPKEtJ1FG_ArW4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1AXbZC25QBPC0PuokhYA5b8WyOR4HvuGr9lZsTR0HMWk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 57600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 57600 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIYn1IItDWJZY2hmD9-0eW5TADd3akO6yu4iNVSScCFLk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 86272 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 86272 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI15qZVVldYFEVxv6DoRCJq1o-8jbrdV19RqRcT-bcTp5g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 86272 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 86272 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1lHcr20MZlAXhwNOR6Gevj4wJaRc4B-tum0bYx49prt8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1XvrvtOqqq8hKEVOCgfEFmALGzVphawbwpkAxG7nqoi8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIYXJbKl66tz5Yz16uScZ-3QcnIi8M-PaVMoZpl5IQJR0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 8] + MIWaveTileA: 4 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 8 + ThreadTileA: 16 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIWMlYxh0gBBwfi1wvTDHJHkSEVXIuF12-m4Tq-srJGFk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 224 + MacroTileA: 128 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 7 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIU-MW4z63wAohwJUDz0C8OcgBozd0DUJEBLSSdararIA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIZLtCnUAcZQRb2Sm7BFv7A3a2vBOCNqxyJPub1dc8Z8U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MInMSU2Gh3mMVemcGZ3WqUWu2l4NG8N5rOCZZl_tZSxr4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1VCOEzWL6LPvmD0HkYK2IQmKeQ33MFQ6q3DD1gUWXPWI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32256 + LdsInitCVgprs: false + LdsNumBytes: 32256 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32256 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1bvrvP40X4hFkSo1KwPg0VEiYDXT6P8LD2PVpHtPr4t8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1KCNveOe_jYnymu00u65Iq4hQQmFEfEECKI4hYCSw86A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 22016 + LdsInitCVgprs: false + LdsNumBytes: 22016 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22016 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI11tDVdUBSEo3UIS5TLyJqHVgQw3dto3agLBzaHthkf5A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 46336 + LdsInitCVgprs: false + LdsNumBytes: 46336 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46336 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1uDCZdlP32w2aUIZQgU6UFaItvoNYlwEGmbTkXc1IEtc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 48384 + LdsInitCVgprs: false + LdsNumBytes: 48384 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48384 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 7] + MIWaveTileA: 3 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 224 + MacroTileA: 96 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 3 + NumLoadsB: 7 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 7 + ThreadTileA: 12 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1tIdq_8HHc1AUHlfryN0_49Q8EAD9oEnK2lMgmhIKC2E= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1o_9BqNra6zh_HmLvWFNQm4WYzN2W1bygUabZEe9Lzsg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1KePfYaceZ29tiGigvOtcCFJNsP7pueMsjzyHKVb7w64= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16tOU8-XLYrNPdvvuRaPZblzPTCq0IYpmOqRVYwuw2zIA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 27904 + LdsInitCVgprs: false + LdsNumBytes: 27904 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27904 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16iQaL89dfDqFaUwHSFuqBVL1d7gxxRqiMEISWe48rBhk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16G697WmrCkkuSW0ygicOEsjSabf2uABvGDaKBNngK4oE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1opMCZdZFzR31a1nXMExXx-HFGwt6FHBNysNBuXPo34o= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1GPhaxX2PStJ8QfX3S-AgQFNJD6o4d_c2qqsOdt_oaC4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 44288 + LdsInitCVgprs: false + LdsNumBytes: 44288 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44288 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 7 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1gqbADNhoP3bHba8oMSfqNYB-76ixvzWU8E58C_oXxBo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1-VJoMMTo08ubflQERsCoO0j_TzMOoFOBMG1VV9dqzpY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI15OARqHSqH8kbGBPru-B8g6Oad-Zo3FGsUS8SxkzZZYk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16xVRyCL8pcYWjhmBX3Sksf7_ECJZKzPUO0YM7WCFd9sM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 23808 + LdsInitCVgprs: false + LdsNumBytes: 23808 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23808 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI162emMCIrLY9uAeJtwZl6HlANVDofIksu3xpgoJdG4L7I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16UNManzuffvMTmsgfWnWF6Bt5jjvtS0tBsdToDQ_bZvc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 24832 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 8448 + LdsOffsetMetadata_Blk: 24832 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1kHn9o1wGnb0Le03rKuL4bApCYLkFlSLtoU6MjkRibzc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1NMWe8o4PG1BIiKP1ggEjZUpwkkJSDyvBBNMZ8tXiaR8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 1 + NumLoadsB: 7 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1Q0rm-ZJ-B1Ep3tDRV7Y6v5xFppIw1pc1wB6W7qCB6eE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 32000 + LdsInitCVgprs: false + LdsNumBytes: 32000 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32000 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI14_XY01zUqMSgVV3Z29zc0HvqY2RBU1XaMixpmKPTFvQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1BV068w-wQamiyACMJrCQO6AzvkFtVVaL6YUOIybq_As= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI167ogjNu2arPxWNgqgKM8sDxj_QbeHe5G3biyDU4kITPE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 19712 + LdsInitCVgprs: false + LdsNumBytes: 19712 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 19712 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI1606uPSrg8jSF_adB8-Q1EUFxmHwf5F9lLMz64s3UmyCk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16Jm6coo6HlXoYc23v3SkS10L-uM9lwa5bnHnY2Vs_oLo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIv9D6qCD557UuGj0WYoB1KLme1Sh1mzoR8zvP-Ew1rDY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 66560 + LdsInitCVgprs: false + LdsNumBytes: 66560 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 66560 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 13] + MIWaveTileA: 4 + MIWaveTileB: 13 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 208 + MacroTileA: 256 + MacroTileB: 208 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 52 + NumLoadsA: 8 + NumLoadsB: 13 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 13 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 13 + ThreadTileA: 16 + ThreadTileB: 13 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIZTzgVMYbF8zAjkLbWPx_KguNjRPGDWzQphgfwu106mI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 28160 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 11] + MIWaveTileA: 4 + MIWaveTileB: 11 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 176 + MacroTileA: 256 + MacroTileB: 176 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 44 + NumLoadsA: 8 + NumLoadsB: 11 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 11 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 11 + ThreadTileA: 16 + ThreadTileB: 11 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIYpXgEPF2KmPhp_HSItkaLdu8ZfTql6wuzRMctpM8bWs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 121856 + LdsInitCVgprs: false + LdsNumBytes: 121856 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 9] + MIWaveTileA: 4 + MIWaveTileB: 9 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 144 + MacroTileA: 256 + MacroTileB: 144 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 8 + NumLoadsB: 9 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 9 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 9 + ThreadTileA: 16 + ThreadTileB: 9 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIM6ky9Tq6VRezat1v-Vy4Qy4OEEUGyFJHwE-XvlIpRaY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 17920 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 112 + MacroTileA: 256 + MacroTileB: 112 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1ZaT28UTzCJBM6wCUWArJk9cu4afCdA6oGhULOQC7jwQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 46080 + LdsInitCVgprs: false + LdsNumBytes: 46080 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 12800 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46080 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1ckLt2U3Q4BSlwSpNR-cpd6uJ3sMr5OVPRWuKvEiSVtY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 7680 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1Cks6yoWu7mRHY5OU_L0wlyt_BkdBZ9B6KYLX0jMjEZM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 35840 + LdsInitCVgprs: false + LdsNumBytes: 35840 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 2560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1LRieWm8obTv5BrugK6ZfaH9J4-zT3LrRBcx6GxwPdUk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 1280 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45312 + LdsInitCVgprs: false + LdsNumBytes: 45312 + LdsNumElementsAlignedA: 10496 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 10496 + LdsOffsetB_Blk: 76032 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45312 + LdsOffsetMetadata_Blk: 76032 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1zBVp9O_j-pDQmFJ0pruVZjPjyXjbaxIDNKuOKzE-hUk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 768 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 6400 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 6400 + LdsOffsetB_Blk: 71936 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 71936 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1LmEbjkFjRe2EHvxZ5njVJSBpnw78dqd37eC05zoq8oY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 2304 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2304 + LdsOffsetB_Blk: 67840 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 67840 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI_Cpsqq7IdCcI6m09ixnRrSz_HTRJbwY41LkpuZ6eaow= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1tP0yLw5sOaqSqLw69T0rUISjIec4gtRcEBnLindXxd0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1WlKSkQYWc45q9-9u3Sa8KR5G5YSwdMXXvYIooja5mzU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1ZFz2Wqi9AGPfO9YZeT8hu0U-a6QRpBHZ1MNZhOuWfzs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16YTIDOkvWpHTvJMbXMK3Qn2TUcHpLDIz3L67zYqwLtxM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1Hyiuad9E_LwogHYOQgHONBkR83eXTUE_-HwHw1z6c2M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 62976 + LdsInitCVgprs: false + LdsNumBytes: 62976 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 62976 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 10 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1_vkhGjNt0PNt_xV8VVOz2XzUCfwEcU8zljVlbKTICVg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16riuLixhIaRRQRAFH27-FM-5iTnYrSUKGh2HFdVNH4W8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16tRLsgniRf9NQ8cwUknJpPeoosk2I0euxrG_yXuRD6Zc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16zCjLhPMzebcvjlA5OUrujTBGN1OYulB3ktj1a570vLs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1OuevmgTEfRQM80_r7hzorlLPNIZiSHCEk0LhjgcOAH8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 60672 + LdsInitCVgprs: false + LdsNumBytes: 60672 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 60672 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 12 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI117HfbCeMXp_75IhCNgyVsAtudVa47Y080duX4tlb_Cc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 54528 + LdsInitCVgprs: false + LdsNumBytes: 54528 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 54528 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 10 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1w_TVfNrjE3mYvq96Bm1Jgkl-HCbXvzNQ6H6gr4PSnKc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16IX89BCLJtYzlNb-bQG51jUiesgW7gn9abBGKzaC4u7Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16yEC0PHvmD0zrKvRDTkhqsmiXq78VvD0Eg5fcAnocoic= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16_OtB8fTrZk3Pt0bE3bXfsWfkoWXEPa2ENPfe4eaiVRI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1uaMQi1863lUOUemwTE99mttXxpTRkvqvmiwvtfx6Qx0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 80384 + LdsInitCVgprs: false + LdsNumBytes: 80384 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 13824 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 80384 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 16 + NumLoadsB: 3 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1_aftRJpk6u1tTs3-Hwls-Fo3A3dD889Os-hWuvfNQvw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 71168 + LdsInitCVgprs: false + LdsNumBytes: 71168 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 4608 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71168 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI18CjcSziK0PjBic2MAopWR-Ta_nHbDse6qNtds5pye7Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 256 + LSPB: 16 + LVCA: 1 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 71936 + LdsInitCVgprs: false + LdsNumBytes: 71936 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 135424 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71936 + LdsOffsetMetadata_Blk: 135424 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 512 + LSPA: 64 + LSPB: 8 + LVCA: 4 + LVCB: 32 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI169OaDNKWZnXMa1R_AlPEgQpVqAFVb54TBTn3PvUMxY7A= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 512 + LSPA: 128 + LSPB: 8 + LVCA: 2 + LVCB: 32 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16JRL6UviIw3ql2MWJlkjDhAPVUQuM-dfo_YP9V8INeCE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16a9rQBxUkyvDWlmJquyIGWqzMINaNpRD_aWL9QWIuzQ8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 512 + LSPA: 128 + LSPB: 8 + LVCA: 2 + LVCB: 32 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16Jyq1ax9kWbpfAk08jNUVf7z1XoXRVCVdpt-kBpa0fWY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 512 + LSPA: 64 + LSPB: 8 + LVCA: 4 + LVCB: 32 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16D0scrs9_FOYkVHUHqfaWIBeu8Rz72d3pOV3R8iV8H74= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI169Ao_eCjaajChoPx9i0y3MGqMyeJdbqU3dbjwB_UIKx4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1DDaFTcewwV5Ybn8swfgQtfH13aPb98dLOnRrTIJYmSQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1ifWE51xli667eSWHdf8X0Gqm1isofKMCZj08QiVl3U0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 1024 + LSPA: 128 + LSPB: 4 + LVCA: 2 + LVCB: 64 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1U06udU6pkFuvB0exyiBHs-wOgMdXOgbcv4nENk4uw44= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1oDw0RAqdNEFhP7idqV0O87totk9ugWGoiXKFb7JRQVw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84992 + LdsInitCVgprs: false + LdsNumBytes: 84992 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84992 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 16 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xqwn4mYe9PqjtRIapRznDsflXKMnaPO3v98RcshyCRoc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 16 + LdsPadB: 8 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index 6dfe08d9e7e1..6b426f926eb7 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: true +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -79,19809 +81,14277 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI93W5TiEccmFg8wGd3V3_i7anSx2in1g2Nvgr1nxFMW0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1WBDB2zkUie8-CEEoI_ArcQdbGzpndP9LCMEirfw4oOM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46336 - LdsInitCVgprs: false - LdsNumBytes: 46336 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46336 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1byiwSEFKpQOHfD3AL3aMNi54Gti0tPrWia0W_RI3F9I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1drE7ijfboWVMJx6XbcD7R5giTiA773IMYwA0SWdbSew= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIlwPgpsfEr1tHJhWc_dQ16xuUglNHBRvb9FY8kgrV2uo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 128 - MacroTileA: 224 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1vpzVkbJ5WB1luwHLUPLp6GT1RJDqsKYFBruB0CuPp3k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48384 - LdsInitCVgprs: false - LdsNumBytes: 48384 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48384 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 3] - MIWaveTileA: 7 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 96 - MacroTileA: 224 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 7 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 3 - ThreadTileA: 28 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1zuV98IyOwxElPoy5ichUD4fa6nssRrdqd1Q3modyHdk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44288 - LdsInitCVgprs: false - LdsNumBytes: 44288 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44288 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 7 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1xqlNtEYE-E0Mz0QATa9_dzumE9-a3Wvr1cR9tzlzYW8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 7 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICgefWmv3HWofIYKXJsq8855MHgkupzgrkVMcs4u-by0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 125952 - LdsInitCVgprs: false - LdsNumBytes: 125952 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 8] - MIWaveTileA: 6 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 256 - MacroTileA: 192 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 8 - ThreadTileA: 24 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MInHiNZ23NKfFsK7oWehXsdVhdC3keTzJy0tZ_ibW9pzc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1rAFgL4PTZQ2gx_CQR3J0YNYjnR6ZpZqLT3kKoVcsb8Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1Vg4WjQQgNmrK6Ytw7ZwOj3TBnXGNte5TZoyB4kGGvxQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1WRxklg9QBBuMW_MrlGv_hyeD_1KjPxvE5CvBb2xGGdk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32000 - LdsInitCVgprs: false - LdsNumBytes: 32000 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 60416 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32000 - LdsOffsetMetadata_Blk: 60416 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MIGmZq9BXDiHJ0tc5zxXOziePTEBPwz6cUZrAUknIRT7E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123904 - LdsInitCVgprs: false - LdsNumBytes: 123904 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 8] - MIWaveTileA: 5 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 256 - MacroTileA: 160 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 160 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 8 - ThreadTileA: 20 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIwjK4OAX-B9dLgo7aLraklJg0ipZLFjGXad70tFlHa_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50432 - LdsInitCVgprs: false - LdsNumBytes: 50432 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50432 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 6] - MIWaveTileA: 5 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 192 - MacroTileA: 160 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 120 - NumLoadsA: 5 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 6 - ThreadTileA: 20 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIXGlL2yJ1nYGKMcvOKU86eOyg9fj72cyU6dwaNXql1aE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1d_LfxY9vu-XLyyQyjblMOjGGccv6SWDTcAMP2nTGrZ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1xJMqskwq5v3yaRA45HS_l-Q0PeuwEvXnjkyc8iEsIuQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1J2YusILFQCwZJTPycybXWTPnKyMth_wvsZJGJUL6XBA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 58368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 58368 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIK3HX_ARTjdrJyQgn8ka1A08cR0NqiFPRF8k7JYx3lno= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50176 - LdsInitCVgprs: false - LdsNumBytes: 50176 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50176 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 8] - MIWaveTileA: 4 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 8 - ThreadTileA: 16 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIAYYyDwfBJqmmCRgY0wORDPvjgSZIA73oaclExRH_3Ks= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIL86z76qmkuTn8D76ItocU1Fqkm1UTu79IZpG8k_14Mo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MIcEWO-iBcRFoqXFm5FnHIJnGhh114ULGYJcbTf6WzdFI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1uj-yswGYT2k5VXIiR3VHictFIePTXmx0STeF7oDnJ6c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1_uypCl737WTznRBFG8qQSTk3KV2VA62EV3JnGAPhybs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1SN9OiW0wmLBhcMdgMSszjYfbUYfIwyxoyCZVswMyYtQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1OdZvyKg_z2Mm3twNkO33DsDFI4XHPhixfFGT8M7-fns= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48128 - LdsInitCVgprs: false - LdsNumBytes: 48128 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48128 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1-rYnfwe4wvrEHrwuVO2V79PzVjbfHc2PuGqRKVJDgfI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1QbCX-NbN9_FulNkEv2WcOQVCSkdBBRKzE6PDyeD0PIc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI15g6oi8kYRfTdSkq4hb7l4vHYA3S6a_b3gQ32PTizD-c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32256 - LdsInitCVgprs: false - LdsNumBytes: 32256 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32256 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16zaeoUJs0j0gHGLtiiAi_mKOg4oYgz97kp3_f949MqD0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 27904 - LdsInitCVgprs: false - LdsNumBytes: 27904 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27904 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16n1I05diAnU09ahBMg4ZOJ69DUU98umJt7phil649P_I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 23808 - LdsInitCVgprs: false - LdsNumBytes: 23808 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23808 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16qYVsRYavL1_uEB81mfehkE0xypBMpwpmv8NTKwWUCZQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 19712 - LdsInitCVgprs: false - LdsNumBytes: 19712 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 19712 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1rLkRPc5Fv6KcZ0pkpjti_dDNkwatHGJAjTt5YgSUF0A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1MsUtatzu8gT7lwtRqOR1GlkYA3LAi1h7-WpiR8njS-s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI17ixososrZhbGip5glmYhzY0dhoWQ6s92NzkMuYpNd6c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1O6oTiyV66Nmoj0i8-AQlJj5R-NLtNsfH6diaQrWbGzQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16BDy2huNg1pYAEVpwRmCSx6TZ3y-fyTBhxwPrXBTUMn0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16T76NxLk5ebc1y9bf0bGaRoLE1Btglvx1amB35nt2jrg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI164mp4kBtFpiXW7TfMmUnYxQd7fjnanNrZue8IY8R1Nt4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29824 - LdsInitCVgprs: false - LdsNumBytes: 29824 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 25600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 9216 - LdsOffsetMetadata_Blk: 25600 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1CBdovKQljtt9JJeN35QUWWaGXoZh_9PeLDfb3HuWMiA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37888 - LdsInitCVgprs: false - LdsNumBytes: 37888 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37888 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1eYnh8ExIG9i1WyAXPWhGKC2q4srWo55-2vcORDH9kEs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1m1tiKe28X_8BSOyU__5gEgsr8TQY2h1rX5gF4hmE4Qc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI17VRJH_5pnmfoUU-FHr32iGMlOdeQUYJplapfD41zqrY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22016 - LdsInitCVgprs: false - LdsNumBytes: 22016 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22016 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16AqtF9WEJ-EpQbhJnOZKbC372wZ4o2kh1rPtJZau1G1U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16WqqFFjCIevU2bTOwvq53GHxOVUs4xuB_iw0OyIMRO9s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16O0pSqsQ7I3lGvs23ES_yKR_wzzq3HSwM1IcyJmGtnhc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25728 - LdsInitCVgprs: false - LdsNumBytes: 25728 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1Jybduy0hX_yQp-cMS6QlHSWe9YueMgbIOjgYTfyE4xY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1280 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45312 - LdsInitCVgprs: false - LdsNumBytes: 45312 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 10496 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45312 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1Mhf-Vi17l2Fosir59CVL3Da6ElGCICe4ioDdRimtZBI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 768 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 6400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1zfbotELeA8YVwjd3HOC5VjWft_ZNaVArBeo8Zgzp6wo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 2304 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIPJDSEwAxzWM4fDuAEI3dit5Qm8ipFSRMSAIZnl6eve4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 66560 - LdsInitCVgprs: false - LdsNumBytes: 66560 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 66560 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [13, 4] - MIWaveTileA: 13 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 208 - MacroTile1: 256 - MacroTileA: 208 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 208 - NumLoadsA: 13 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 13 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 52 - ThreadTile1: 4 - ThreadTileA: 52 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIHlGj0Igh78VESckoEVhEEojHcEyAVYdCerN_cX6Ldrw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 28160 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 28160 - LdsOffsetB_Blk: 93696 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 28160 - LdsOffsetMetadata_Blk: 93696 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [11, 4] - MIWaveTileA: 11 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 176 - MacroTile1: 256 - MacroTileA: 176 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 176 - NumLoadsA: 11 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 11 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 44 - ThreadTile1: 4 - ThreadTileA: 44 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MIs43vjb8vXLGfeBgy4ohTkGSxXW8PsCp3uz_zeqpVxeM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 121856 - LdsInitCVgprs: false - LdsNumBytes: 121856 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 88576 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23040 - LdsOffsetMetadata_Blk: 88576 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [9, 4] - MIWaveTileA: 9 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 144 - MacroTile1: 256 - MacroTileA: 144 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 144 - NumLoadsA: 9 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 9 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 36 - ThreadTile1: 4 - ThreadTileA: 36 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIYY998OJmuRcBnYy6KGEItwDq2j2QehUOs8VfioGt_EY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17920 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17920 - LdsOffsetB_Blk: 83456 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 83456 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 112 - MacroTile1: 256 - MacroTileA: 112 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1riWq2NrSVGbHY4ulT3CBZ1ka9jZSPImqGDsOLfR739U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46080 - LdsInitCVgprs: false - LdsNumBytes: 46080 - LdsNumElementsAlignedA: 12800 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12800 - LdsOffsetB_Blk: 78336 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46080 - LdsOffsetMetadata_Blk: 78336 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI109tIWmrNFpIMCvCKD44FEjLK6mNrPP0ecMYRKJWv__o= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 7680 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 7680 - LdsOffsetB_Blk: 73216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 73216 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1ePlkOMYOjA6Sj3U5aK9R-vdkoYkQzZKOf49wW2bLE70= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35840 - LdsInitCVgprs: false - LdsNumBytes: 35840 - LdsNumElementsAlignedA: 2560 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2560 - LdsOffsetB_Blk: 68096 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 68096 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1R3pW5TGY8A5opfshY0CC3g5uD4DXiGRwS1VnQHSdtTs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 60672 - LdsInitCVgprs: false - LdsNumBytes: 60672 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 117760 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 60672 - LdsOffsetMetadata_Blk: 117760 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI13Sbf9PQEx5F-iKE9julI7_PdDt3yPapFP4vCFUpDo1E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 62976 - LdsInitCVgprs: false - LdsNumBytes: 62976 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 62976 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 10 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI137t5Uvc0ssl50sE4SNlCcUp-XUoCcLX9_I4uGMXledk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 54528 - LdsInitCVgprs: false - LdsNumBytes: 54528 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 54528 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 10 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIOmRHwGdcC8gFOr3dPzasFfQWs0PP2cDcwy6aKZU2f5U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1454Bwxd_MIvQjuG-Y_CvHs_qnR7PgRRvHGhCUO-XbOY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1Rmqg4myY3vDAMf_l0hT_13f8mLa__KusFSpIvX00Gq4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI15NEBMkxG3p9Ig0Cr9U7Vscm7bIb_z6qr2ylZCFd2y30= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16wxJkzA64nljgT6A4tMYPR-qnvL3NK7-LGLeTrL93K28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16s4XHH5Mm_pX3J5f5thK2zPFz4wk3wfVK1ZsVsKy_LpE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1XeU_kq-TCV7gFRA0zjadiPg5xVzK4J_9tcfhp6RHcE8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16uy_AuCnsXVT_W8ZUwPY2xZhbdvr0s1-k6m6FDS2GyPc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16JHWtGGP8udHdAtFC_ITt55YeNZjw7qy9YbaDQmBmLnw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1CMycMeJchCKewZBbLPkfdDttQ7jpxPNi2DcgEi0aA7Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI167PfjxLOUHYcmsWqM3Jr5QHvep_HsB981kR15bVmyR_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16hW4G8jnTFUiWZWnW0Tmfmi9zpySUXkWHngzRijqsswA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16pkwZPXEjbcu5SfItYZgOkhxquryDik7C3Mlw6DhdC9c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1ag0qOUjDoYcyoioR4OdNEh8L5HMCVTkf-1kVLG0Ke8Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 256 - LVCA: 16 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71936 - LdsInitCVgprs: false - LdsNumBytes: 71936 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71936 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1RElBQ1lA5tfxLjjMGZY4_t9WElkqvIrWw-ZvypHxsS4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 80384 - LdsInitCVgprs: false - LdsNumBytes: 80384 - LdsNumElementsAlignedA: 13824 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 13824 - LdsOffsetB_Blk: 144896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 80384 - LdsOffsetMetadata_Blk: 144896 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1g4O7XQYq5Erjhq0mbnd0ZzsOn7Y74NZ0Ed50NCU8Cko= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71168 - LdsInitCVgprs: false - LdsNumBytes: 71168 - LdsNumElementsAlignedA: 4608 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4608 - LdsOffsetB_Blk: 135680 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71168 - LdsOffsetMetadata_Blk: 135680 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 64 - LSPA: 8 - LSPB: 64 - LVCA: 32 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16c-3ymAqE5iIzAVIl4M4mFas_yKBzwB8S5hIJHrx_G10= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 32 - LSPA: 8 - LSPB: 128 - LVCA: 32 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16l2koiE9edxgF5EcPYxeXQhdktTUf3O_Nh5I8-rOv_5c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16PDTjUmNdBPeXIoutsSh54JIRkDgOgTuGJ5_UUnXYC30= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16FZJXxZzBUYpkHY1uHTOo5xfvmHSftJ7Ss8JFmsnHOq4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16UExxzleg69jKCOIPv3aHKRuxpeCJ0OtN8U8_i8380wE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 32 - LSPA: 8 - LSPB: 128 - LVCA: 32 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16486QKgKm2Q-ueCL9nw3616pMoSPYJwy96wpaKy2ZR9g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 64 - LSPA: 8 - LSPB: 64 - LVCA: 32 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1gJ-iaCReTZwgtOaq-r_7Ae1Xx5lpHHILXpRm4hXIFWo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1TOymj1f06yNxQYQNnDElBwV9Je3kDkNSQdL4ZFlGHxw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1eVyecsmYTDI_tnvIi0T4aak3HoYfyCs1_BMjsWhSVGA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84992 - LdsInitCVgprs: false - LdsNumBytes: 84992 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84992 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1SvhFCnFpOT5VXBaOtUT022uRF_6hTz1blD7UqHf2kpA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 32 - LSPA: 4 - LSPB: 128 - LVCA: 64 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xDWExf5YqS6e0I09q-b7LtSPBEclPiz1pvMgNxDpD5yY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 8 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI93W5TiEccmFg8wGd3V3_i7anSx2in1g2Nvgr1nxFMW0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1WBDB2zkUie8-CEEoI_ArcQdbGzpndP9LCMEirfw4oOM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 46336 + LdsInitCVgprs: false + LdsNumBytes: 46336 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46336 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1byiwSEFKpQOHfD3AL3aMNi54Gti0tPrWia0W_RI3F9I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1drE7ijfboWVMJx6XbcD7R5giTiA773IMYwA0SWdbSew= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIlwPgpsfEr1tHJhWc_dQ16xuUglNHBRvb9FY8kgrV2uo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 128 + MacroTileA: 224 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 4 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1vpzVkbJ5WB1luwHLUPLp6GT1RJDqsKYFBruB0CuPp3k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 48384 + LdsInitCVgprs: false + LdsNumBytes: 48384 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48384 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 3] + MIWaveTileA: 7 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 96 + MacroTileA: 224 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 7 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 3 + ThreadTileA: 28 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1zuV98IyOwxElPoy5ichUD4fa6nssRrdqd1Q3modyHdk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 44288 + LdsInitCVgprs: false + LdsNumBytes: 44288 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44288 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 7 + NumLoadsB: 2 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1xqlNtEYE-E0Mz0QATa9_dzumE9-a3Wvr1cR9tzlzYW8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 7 + NumLoadsB: 1 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICgefWmv3HWofIYKXJsq8855MHgkupzgrkVMcs4u-by0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 125952 + LdsInitCVgprs: false + LdsNumBytes: 125952 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 8] + MIWaveTileA: 6 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 256 + MacroTileA: 192 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 8 + ThreadTileA: 24 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MInHiNZ23NKfFsK7oWehXsdVhdC3keTzJy0tZ_ibW9pzc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1rAFgL4PTZQ2gx_CQR3J0YNYjnR6ZpZqLT3kKoVcsb8Y= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1Vg4WjQQgNmrK6Ytw7ZwOj3TBnXGNte5TZoyB4kGGvxQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1WRxklg9QBBuMW_MrlGv_hyeD_1KjPxvE5CvBb2xGGdk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 32000 + LdsInitCVgprs: false + LdsNumBytes: 32000 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 60416 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32000 + LdsOffsetMetadata_Blk: 60416 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MIGmZq9BXDiHJ0tc5zxXOziePTEBPwz6cUZrAUknIRT7E= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 123904 + LdsInitCVgprs: false + LdsNumBytes: 123904 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 8] + MIWaveTileA: 5 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 256 + MacroTileA: 160 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 160 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 8 + ThreadTileA: 20 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIwjK4OAX-B9dLgo7aLraklJg0ipZLFjGXad70tFlHa_4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 50432 + LdsInitCVgprs: false + LdsNumBytes: 50432 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50432 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 6] + MIWaveTileA: 5 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 192 + MacroTileA: 160 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 120 + NumLoadsA: 5 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 6 + ThreadTileA: 20 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIXGlL2yJ1nYGKMcvOKU86eOyg9fj72cyU6dwaNXql1aE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1d_LfxY9vu-XLyyQyjblMOjGGccv6SWDTcAMP2nTGrZ4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1xJMqskwq5v3yaRA45HS_l-Q0PeuwEvXnjkyc8iEsIuQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1J2YusILFQCwZJTPycybXWTPnKyMth_wvsZJGJUL6XBA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 58368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 58368 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIK3HX_ARTjdrJyQgn8ka1A08cR0NqiFPRF8k7JYx3lno= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 50176 + LdsInitCVgprs: false + LdsNumBytes: 50176 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50176 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 8] + MIWaveTileA: 4 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 8 + ThreadTileA: 16 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIAYYyDwfBJqmmCRgY0wORDPvjgSZIA73oaclExRH_3Ks= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIL86z76qmkuTn8D76ItocU1Fqkm1UTu79IZpG8k_14Mo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MIcEWO-iBcRFoqXFm5FnHIJnGhh114ULGYJcbTf6WzdFI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1uj-yswGYT2k5VXIiR3VHictFIePTXmx0STeF7oDnJ6c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1_uypCl737WTznRBFG8qQSTk3KV2VA62EV3JnGAPhybs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1SN9OiW0wmLBhcMdgMSszjYfbUYfIwyxoyCZVswMyYtQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1OdZvyKg_z2Mm3twNkO33DsDFI4XHPhixfFGT8M7-fns= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 48128 + LdsInitCVgprs: false + LdsNumBytes: 48128 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48128 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1-rYnfwe4wvrEHrwuVO2V79PzVjbfHc2PuGqRKVJDgfI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1QbCX-NbN9_FulNkEv2WcOQVCSkdBBRKzE6PDyeD0PIc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI15g6oi8kYRfTdSkq4hb7l4vHYA3S6a_b3gQ32PTizD-c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 32256 + LdsInitCVgprs: false + LdsNumBytes: 32256 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32256 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16zaeoUJs0j0gHGLtiiAi_mKOg4oYgz97kp3_f949MqD0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 27904 + LdsInitCVgprs: false + LdsNumBytes: 27904 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27904 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16n1I05diAnU09ahBMg4ZOJ69DUU98umJt7phil649P_I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 23808 + LdsInitCVgprs: false + LdsNumBytes: 23808 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23808 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16qYVsRYavL1_uEB81mfehkE0xypBMpwpmv8NTKwWUCZQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 19712 + LdsInitCVgprs: false + LdsNumBytes: 19712 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 19712 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1rLkRPc5Fv6KcZ0pkpjti_dDNkwatHGJAjTt5YgSUF0A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1MsUtatzu8gT7lwtRqOR1GlkYA3LAi1h7-WpiR8njS-s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI17ixososrZhbGip5glmYhzY0dhoWQ6s92NzkMuYpNd6c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1O6oTiyV66Nmoj0i8-AQlJj5R-NLtNsfH6diaQrWbGzQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16BDy2huNg1pYAEVpwRmCSx6TZ3y-fyTBhxwPrXBTUMn0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16T76NxLk5ebc1y9bf0bGaRoLE1Btglvx1amB35nt2jrg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI164mp4kBtFpiXW7TfMmUnYxQd7fjnanNrZue8IY8R1Nt4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29824 + LdsInitCVgprs: false + LdsNumBytes: 29824 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 25600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 9216 + LdsOffsetMetadata_Blk: 25600 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1CBdovKQljtt9JJeN35QUWWaGXoZh_9PeLDfb3HuWMiA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 37888 + LdsInitCVgprs: false + LdsNumBytes: 37888 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37888 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1eYnh8ExIG9i1WyAXPWhGKC2q4srWo55-2vcORDH9kEs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1m1tiKe28X_8BSOyU__5gEgsr8TQY2h1rX5gF4hmE4Qc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI17VRJH_5pnmfoUU-FHr32iGMlOdeQUYJplapfD41zqrY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 22016 + LdsInitCVgprs: false + LdsNumBytes: 22016 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22016 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16AqtF9WEJ-EpQbhJnOZKbC372wZ4o2kh1rPtJZau1G1U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16WqqFFjCIevU2bTOwvq53GHxOVUs4xuB_iw0OyIMRO9s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16O0pSqsQ7I3lGvs23ES_yKR_wzzq3HSwM1IcyJmGtnhc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25728 + LdsInitCVgprs: false + LdsNumBytes: 25728 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1Jybduy0hX_yQp-cMS6QlHSWe9YueMgbIOjgYTfyE4xY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1280 + LdsBytesNoAmax: 45312 + LdsInitCVgprs: false + LdsNumBytes: 45312 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 10496 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45312 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1Mhf-Vi17l2Fosir59CVL3Da6ElGCICe4ioDdRimtZBI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 768 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 6400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1zfbotELeA8YVwjd3HOC5VjWft_ZNaVArBeo8Zgzp6wo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 2304 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIPJDSEwAxzWM4fDuAEI3dit5Qm8ipFSRMSAIZnl6eve4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 66560 + LdsInitCVgprs: false + LdsNumBytes: 66560 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 66560 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [13, 4] + MIWaveTileA: 13 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 208 + MacroTile1: 256 + MacroTileA: 208 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 208 + NumLoadsA: 13 + NumLoadsB: 8 + NumLoadsPerpendicularA: 13 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 52 + ThreadTile1: 4 + ThreadTileA: 52 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIHlGj0Igh78VESckoEVhEEojHcEyAVYdCerN_cX6Ldrw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 28160 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 28160 + LdsOffsetB_Blk: 93696 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 28160 + LdsOffsetMetadata_Blk: 93696 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [11, 4] + MIWaveTileA: 11 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 176 + MacroTile1: 256 + MacroTileA: 176 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 176 + NumLoadsA: 11 + NumLoadsB: 8 + NumLoadsPerpendicularA: 11 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 44 + ThreadTile1: 4 + ThreadTileA: 44 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MIs43vjb8vXLGfeBgy4ohTkGSxXW8PsCp3uz_zeqpVxeM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 121856 + LdsInitCVgprs: false + LdsNumBytes: 121856 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 88576 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23040 + LdsOffsetMetadata_Blk: 88576 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [9, 4] + MIWaveTileA: 9 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 144 + MacroTile1: 256 + MacroTileA: 144 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 144 + NumLoadsA: 9 + NumLoadsB: 8 + NumLoadsPerpendicularA: 9 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 36 + ThreadTile1: 4 + ThreadTileA: 36 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIYY998OJmuRcBnYy6KGEItwDq2j2QehUOs8VfioGt_EY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17920 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17920 + LdsOffsetB_Blk: 83456 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 83456 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 112 + MacroTile1: 256 + MacroTileA: 112 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1riWq2NrSVGbHY4ulT3CBZ1ka9jZSPImqGDsOLfR739U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 46080 + LdsInitCVgprs: false + LdsNumBytes: 46080 + LdsNumElementsAlignedA: 12800 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12800 + LdsOffsetB_Blk: 78336 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46080 + LdsOffsetMetadata_Blk: 78336 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI109tIWmrNFpIMCvCKD44FEjLK6mNrPP0ecMYRKJWv__o= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 7680 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 7680 + LdsOffsetB_Blk: 73216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 73216 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1ePlkOMYOjA6Sj3U5aK9R-vdkoYkQzZKOf49wW2bLE70= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 35840 + LdsInitCVgprs: false + LdsNumBytes: 35840 + LdsNumElementsAlignedA: 2560 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2560 + LdsOffsetB_Blk: 68096 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 68096 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1R3pW5TGY8A5opfshY0CC3g5uD4DXiGRwS1VnQHSdtTs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 60672 + LdsInitCVgprs: false + LdsNumBytes: 60672 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 117760 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 60672 + LdsOffsetMetadata_Blk: 117760 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 2 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI13Sbf9PQEx5F-iKE9julI7_PdDt3yPapFP4vCFUpDo1E= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 62976 + LdsInitCVgprs: false + LdsNumBytes: 62976 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 62976 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 10 + NumLoadsB: 4 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI137t5Uvc0ssl50sE4SNlCcUp-XUoCcLX9_I4uGMXledk= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 54528 + LdsInitCVgprs: false + LdsNumBytes: 54528 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 54528 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 10 + NumLoadsB: 2 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIOmRHwGdcC8gFOr3dPzasFfQWs0PP2cDcwy6aKZU2f5U= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1454Bwxd_MIvQjuG-Y_CvHs_qnR7PgRRvHGhCUO-XbOY= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1Rmqg4myY3vDAMf_l0hT_13f8mLa__KusFSpIvX00Gq4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI15NEBMkxG3p9Ig0Cr9U7Vscm7bIb_z6qr2ylZCFd2y30= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16wxJkzA64nljgT6A4tMYPR-qnvL3NK7-LGLeTrL93K28= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16s4XHH5Mm_pX3J5f5thK2zPFz4wk3wfVK1ZsVsKy_LpE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1XeU_kq-TCV7gFRA0zjadiPg5xVzK4J_9tcfhp6RHcE8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16uy_AuCnsXVT_W8ZUwPY2xZhbdvr0s1-k6m6FDS2GyPc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16JHWtGGP8udHdAtFC_ITt55YeNZjw7qy9YbaDQmBmLnw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1CMycMeJchCKewZBbLPkfdDttQ7jpxPNi2DcgEi0aA7Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI167PfjxLOUHYcmsWqM3Jr5QHvep_HsB981kR15bVmyR_4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16hW4G8jnTFUiWZWnW0Tmfmi9zpySUXkWHngzRijqsswA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16pkwZPXEjbcu5SfItYZgOkhxquryDik7C3Mlw6DhdC9c= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1ag0qOUjDoYcyoioR4OdNEh8L5HMCVTkf-1kVLG0Ke8Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 256 + LVCA: 16 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 71936 + LdsInitCVgprs: false + LdsNumBytes: 71936 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71936 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1RElBQ1lA5tfxLjjMGZY4_t9WElkqvIrWw-ZvypHxsS4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 80384 + LdsInitCVgprs: false + LdsNumBytes: 80384 + LdsNumElementsAlignedA: 13824 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 13824 + LdsOffsetB_Blk: 144896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 80384 + LdsOffsetMetadata_Blk: 144896 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 16 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1g4O7XQYq5Erjhq0mbnd0ZzsOn7Y74NZ0Ed50NCU8Cko= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 71168 + LdsInitCVgprs: false + LdsNumBytes: 71168 + LdsNumElementsAlignedA: 4608 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4608 + LdsOffsetB_Blk: 135680 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71168 + LdsOffsetMetadata_Blk: 135680 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 64 + LSPA: 8 + LSPB: 64 + LVCA: 32 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16c-3ymAqE5iIzAVIl4M4mFas_yKBzwB8S5hIJHrx_G10= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 32 + LSPA: 8 + LSPB: 128 + LVCA: 32 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16l2koiE9edxgF5EcPYxeXQhdktTUf3O_Nh5I8-rOv_5c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16PDTjUmNdBPeXIoutsSh54JIRkDgOgTuGJ5_UUnXYC30= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16FZJXxZzBUYpkHY1uHTOo5xfvmHSftJ7Ss8JFmsnHOq4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16UExxzleg69jKCOIPv3aHKRuxpeCJ0OtN8U8_i8380wE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 32 + LSPA: 8 + LSPB: 128 + LVCA: 32 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16486QKgKm2Q-ueCL9nw3616pMoSPYJwy96wpaKy2ZR9g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 64 + LSPA: 8 + LSPB: 64 + LVCA: 32 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1gJ-iaCReTZwgtOaq-r_7Ae1Xx5lpHHILXpRm4hXIFWo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1TOymj1f06yNxQYQNnDElBwV9Je3kDkNSQdL4ZFlGHxw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1eVyecsmYTDI_tnvIi0T4aak3HoYfyCs1_BMjsWhSVGA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 84992 + LdsInitCVgprs: false + LdsNumBytes: 84992 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84992 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 4 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1SvhFCnFpOT5VXBaOtUT022uRF_6hTz1blD7UqHf2kpA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 32 + LSPA: 4 + LSPB: 128 + LVCA: 64 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM1_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xDWExf5YqS6e0I09q-b7LtSPBEclPiz1pvMgNxDpD5yY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 8 + LdsPadB: 16 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index b2bdee0c482f..65d670e8424a 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/Origami_ntb4/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: true +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -79,31283 +81,22382 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MIrZs5mdCNmpikVNpV2YQQNFBCAdivVS8D1AsBXrhUP4k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 8] - MIWaveTileA: 8 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 256 - MacroTileA: 256 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 256 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 8 - ThreadTileA: 32 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MIADOZHpI0e4kz3PekZsq-nbyBZU7EZIxEg7KcvUC6g3g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 7] - MIWaveTileA: 8 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 224 - MacroTileA: 256 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 224 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 7 - ThreadTileA: 32 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIszVB7i9sCthoswarzra0qLsMXdGUHQA9HKGzu-1z5Kk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 6] - MIWaveTileA: 8 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 192 - MacroTileA: 256 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 6 - ThreadTileA: 32 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI4oZDINgtVU-f19tz2HYWh_LmdGFiZplMuFYBiLKIRVE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 124928 - LdsInitCVgprs: false - LdsNumBytes: 124928 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 5] - MIWaveTileA: 8 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 160 - MacroTileA: 256 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 5 - ThreadTileA: 32 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIQveu_FRPQ7kB2acKv85tmm0lWhIAYF1e-bVgSb8HhRU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1nNTR_ZcKmsj4A34uOMZTZX_oilYaQuaQFL2SNhs8cXo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 49152 - LdsInitCVgprs: false - LdsNumBytes: 49152 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 49152 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI17E3Elp9R2iGRIjoSUJt7ENU0IkIEDCYD27G5goO_zfQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI14oL9i2wAQa7K6yEREmecBQB_z40_9_5z5qamIYTi9BA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38912 - LdsInitCVgprs: false - LdsNumBytes: 38912 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38912 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MIVqWf-zHb4fGrqaI_PNfaMIuBSbf24WaOtDVKEmRwTy0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 166912 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 166912 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 8] - MIWaveTileA: 7 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 256 - MacroTileA: 224 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 224 - NumGlobalWriteVectorsPerThread: 224 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 8 - ThreadTileA: 28 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI2FE1_xdi6gisBeCfrvgksfzzbx9Yx6-LxacJqFoorOs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71680 - LdsInitCVgprs: false - LdsNumBytes: 71680 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 166912 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71680 - LdsOffsetMetadata_Blk: 166912 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 7] - MIWaveTileA: 7 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 224 - MacroTileA: 224 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 196 - NumGlobalWriteVectorsPerThread: 196 - NumLoadsA: 7 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 7 - ThreadTileA: 28 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MItgCKUscKBBiMtBOmdwUt6qAHeM_MR7YbKWEkh39p8UQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 129024 - LdsInitCVgprs: false - LdsNumBytes: 129024 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 6] - MIWaveTileA: 7 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 192 - MacroTileA: 224 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 168 - NumGlobalWriteVectorsPerThread: 168 - NumLoadsA: 7 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 6 - ThreadTileA: 28 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI_UOU6HZrGnyaniPidkPyA2vc3eMgOLn6XHhdMyr3gok= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 5] - MIWaveTileA: 7 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 160 - MacroTileA: 224 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 140 - NumGlobalWriteVectorsPerThread: 140 - NumLoadsA: 7 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 5 - ThreadTileA: 28 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIdhuZzZrmslv6fIdikHYnhn1dpXNXnkfYYXLJoFZzt6s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 128 - MacroTileA: 224 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1b5RoEaxogDkty9pH0a4h4iqEK74VHsxBERA-H-NeUNY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 3] - MIWaveTileA: 7 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 96 - MacroTileA: 224 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 7 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 3 - ThreadTileA: 28 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1g3lxPhwHBbhj-5_vMz-ezv_Gct2IZefQGw3UsfUuGuM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 7 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1pRHgvlAR7N2hth7GOj12UHkPW1O97Tnl14UAnpJSv9c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 7 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICt7le9q16ngZmPo_ZNYkoWOjOg2jVjonbKgLeTEzLxQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 8] - MIWaveTileA: 6 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 256 - MacroTileA: 192 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 8 - ThreadTileA: 24 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MIFCEugnbAE1o_KNOxQTjmgk1-riRp7-bqr1XYLHQkDJI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 129024 - LdsInitCVgprs: false - LdsNumBytes: 129024 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 7] - MIWaveTileA: 6 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 224 - MacroTileA: 192 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 168 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 6 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 7 - ThreadTileA: 24 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MIsXAzSnuik23-u_Tqj9IA9yIVlTTHvUnKVAh5tdPfE-0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 120832 - LdsInitCVgprs: false - LdsNumBytes: 120832 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 6] - MIWaveTileA: 6 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 192 - MacroTileA: 192 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 6 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 6 - ThreadTileA: 24 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIdawCsxOV8Q1yfZkDghTj5ROObRZEHHk5r8TLSJ8-4DI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 5] - MIWaveTileA: 6 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 160 - MacroTileA: 192 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 5 - ThreadTileA: 24 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI96PVzcjbGKsmGAPHOizb5QovZZPMnBXqYWXYWCMR6RE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1XF9bGkyS2smPrKpZKRvXqZE_IlG80gFUg26F5mTZLis= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1cyGRGs5fNOjzNULtz2NEdDFcpvgD-53w5wck9Mz0rEA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1dXOVrMlvBwO6y7ESILkUB5p9GtB-Km9ykpVzg9YqhsY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 60416 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 60416 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI_FMKfIaIlw04mnTidFqT-Fnj0gDbPBKMLBy9AxGoipc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 124928 - LdsInitCVgprs: false - LdsNumBytes: 124928 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 8] - MIWaveTileA: 5 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 256 - MacroTileA: 160 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 160 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 8 - ThreadTileA: 20 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MIgvcBHckHBXx7rYEXvd58PMUeCwZSGsDJtMgKSLGOngQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 7] - MIWaveTileA: 5 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 224 - MacroTileA: 160 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 140 - NumGlobalWriteVectorsPerThread: 140 - NumLoadsA: 5 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 7 - ThreadTileA: 20 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIiC0cBFJQlkymU09JdEq7dUq1wvcSX9wghKOSQxkOErI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 6] - MIWaveTileA: 5 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 192 - MacroTileA: 160 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 120 - NumLoadsA: 5 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 6 - ThreadTileA: 20 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MIwREMu9dXpHu0hNi1Te8Ryumj0HYpn1BFkusQoNfhAQ8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 5] - MIWaveTileA: 5 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 160 - MacroTileA: 160 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 100 - NumGlobalWriteVectorsPerThread: 100 - NumLoadsA: 5 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 5 - ThreadTileA: 20 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIcrglZoxFbHHjwhvMHFjwAoFdYaccIqWMvEsbqy0yalI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1BUsuevlFp3qE9qBkCpl8ZHtyZp2nTMvZX1nz1cQRT7M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1IVM_kq6fQOztyL87pw0KKbmpiglUkaPXmejIGutUrh4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1bgH1ukcWIV-QEjZUrEMgbOOF-CIy8Yu_mpQb3L9Zq3g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 58368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 58368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51712 - LdsInitCVgprs: false - LdsNumBytes: 51712 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51712 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [32, 32, 64, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 64 - MatrixInstM: 32 - MatrixInstN: 32 - MatrixInstruction: [32, 32, 64, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 4 - NonTemporalD: 4 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 16 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU8_SUM1_SUS256_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM8_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 8 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: true - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 4 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 2 - SubGroup1: 128 - SubGroupA: 2 - SubGroupB: 128 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 64 - ThreadTile1: 2 - ThreadTileA: 64 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 1 - enableLDSTrA: 0 - enableLDSTrB: 0 - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIHAxeqAl5GZyYpWyj9YBlHLiAot0B1CVx9Nf_5OHRC1Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 224 - MacroTileA: 128 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIIlrnP4n1WnSi3fe0xoxepRfTuuLMMp-jgbnYdGnnvYE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI5bZ7yznefmkxpzzbB-XiEKlKRJu2AMeEgvXx60Tq5L0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI72iEPgzvyM2ZrWa9mJE7BjRqJtpIrgM81I_tGNEjK0c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1xf_ukgQ-Z-qNS4jmkabhCAr3mB7WAQFBVo0jxB_s6wM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1j8uVA7i_pGx8ZCBrMR8ZhK3U6bp7J7YjtbYnnKdPxDI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1M6uR2wC2h6sr2Hl9YrF8zhWGQ7RgoVsQ-TM7lCE6pPI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22528 - LdsInitCVgprs: false - LdsNumBytes: 22528 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22528 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1lw4esJJcKXjrN7F_gKTEASvr4FPioQKVPOZImphrvBY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 49152 - LdsInitCVgprs: false - LdsNumBytes: 49152 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 49152 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1JixbiZckG_HlXVPjjWZHNInPUwLf2xyuU7BTFa5yHI0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 7] - MIWaveTileA: 3 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 224 - MacroTileA: 96 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 3 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 7 - ThreadTileA: 12 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1wIqbeUYowBvH47eY0D90voHkVaoBg2lPdbX2d68nIJ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1J-2M-3q4LSCntZjUh3v8nZH3qUcDC89rCJ0Fdq3iMeI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1Vr1CG6o8iRf8BbhZJxSu0FVBXoX31LtlPuM8cs_YEXU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16Cnahg_B3KmoGxZkTO1OYZLBTHOrwHNY1VQd2t5IrylI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16W4yMmcRT_Rcp3ZNo2gSW3q4SJdd9i9DRRKcgqox59VE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24576 - LdsInitCVgprs: false - LdsNumBytes: 24576 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 24576 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16zNms4_iu6xgFQiq42WUbXrNnFNyslp04sbAKWAlTbog= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20480 - LdsInitCVgprs: false - LdsNumBytes: 20480 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20480 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI11aFcbk23grO4-hK8aBkA0Q2TJ_RzSnSefhCsshDFWE8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1BIQ9uHrC9E8FHJ9mzReo0log_q-mFe47Gvfw9lRTxKU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1oWIDEj_bU2sIdW6WZSqKJI8Okx_mH0MwWvgKg4p0Wh0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1DfS6ucItLGFYy3q0Zz-JZd7icG_ycHgazunawsT8Pns= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1jK-MUAAqzxLFJYDxWtp_lRnOH6G_Sa3LS7gsDkzJ5ZI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16qBSTIiXAL1b2k0UyETAi0pu4GSfWwKXX5o-ybzBtncE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24576 - LdsInitCVgprs: false - LdsNumBytes: 24576 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 24576 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16mOGETtdY_W1srxs4Mfgd04W-cV79aCR6raMD3HA6jWU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 18432 - LdsInitCVgprs: false - LdsNumBytes: 18432 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 18432 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16wIclDOa00a_BcP3sOUDpDc_qFY9RvbDE777DV1a9O7k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 25600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 9216 - LdsOffsetMetadata_Blk: 25600 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1pg1aJvOuqnmTxe_-2e9_1S4uFGoAaPhgR-zs28ui024= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38912 - LdsInitCVgprs: false - LdsNumBytes: 38912 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38912 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1bJJ5iziycwP81Vg_rFbnRobq-WE2Gw-XuD7kvGV58LU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 1 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1c60kFRwy-kwCbpVB3Be-Tp-X0igfgGU9wMtdCOAL5xM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI13_eugo-sWMKarMTBe1WI4hmzJdiGvToykcHBbj4WXow= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1gSIhM4a60TQQmpGca3zPH_1C9YMOrBSb2Bu9JrxvD-k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22528 - LdsInitCVgprs: false - LdsNumBytes: 22528 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22528 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16YH7g6N9l5w5imlP3Lp9mdDXwtO_ZtzNO9yUXMFVOjKQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20480 - LdsInitCVgprs: false - LdsNumBytes: 20480 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20480 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16IFZMDGDs3-q6koIU2xIB_XUAfzhmng9Mg_46GO3Rqf4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16rLKaVqye5ZVlglTUO9ioHrZ6WB-GJ_Qs2RqP4S8ZIXE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MIyoGz0G9DWmdNz6WjS2Bq6rCsCrR_mXM4td8eMkMYEDw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73216 - LdsInitCVgprs: false - LdsNumBytes: 73216 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 38400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 165888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73216 - LdsOffsetMetadata_Blk: 165888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 15] - MIWaveTileA: 4 - MIWaveTileB: 15 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 240 - MacroTileA: 256 - MacroTileB: 240 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 240 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 8 - NumLoadsB: 15 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 15 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 15 - ThreadTileA: 16 - ThreadTileB: 15 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIBnc7kWhy6AvjwbNc5adjziCfJiLAVAD1RzoNcS6OMZQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 68096 - LdsInitCVgprs: false - LdsNumBytes: 68096 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 165888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 68096 - LdsOffsetMetadata_Blk: 165888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 13] - MIWaveTileA: 4 - MIWaveTileB: 13 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 208 - MacroTileA: 256 - MacroTileB: 208 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 52 - NumLoadsA: 8 - NumLoadsB: 13 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 13 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 13 - ThreadTileA: 16 - ThreadTileB: 13 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIF5A-1D1ZIPjJsJJ7ttt4AlfEFPQG0riur7Zv_qXG5YI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 128512 - LdsInitCVgprs: false - LdsNumBytes: 128512 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 28160 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 11] - MIWaveTileA: 4 - MIWaveTileB: 11 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 176 - MacroTileA: 256 - MacroTileB: 176 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 44 - NumLoadsA: 8 - NumLoadsB: 11 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 11 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 11 - ThreadTileA: 16 - ThreadTileB: 11 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIyRCs1CYgk6qoLPVzZz2698DWR4wMJ7whNVJ8JwJfnZc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123392 - LdsInitCVgprs: false - LdsNumBytes: 123392 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 9] - MIWaveTileA: 4 - MIWaveTileB: 9 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 144 - MacroTileA: 256 - MacroTileB: 144 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 8 - NumLoadsB: 9 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 9 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 9 - ThreadTileA: 16 - ThreadTileB: 9 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIaVDafPaNDXaLtu56b1YkZsQUI1aLmxb3jgdu_dv65EA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 17920 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 112 - MacroTileA: 256 - MacroTileB: 112 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1jNVYmoC6fjSjbIADTf0f9l0Cs4TKifqjd6Xql0ehllk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 47616 - LdsInitCVgprs: false - LdsNumBytes: 47616 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 12800 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 47616 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1eX_SljyDEcS5rCPMhk2caJwb3hhe7qqzmb_HRJ0i-U0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 7680 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1fPZ5I9MI2WZKscvBb8L9C1UGdH6oze92UNxsd_rmjZ0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37376 - LdsInitCVgprs: false - LdsNumBytes: 37376 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 2560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37376 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI17whl_3k4Y1EhiqZzn7y-o-UTqlat-Sd3AsiBr6V3nc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73216 - LdsInitCVgprs: false - LdsNumBytes: 73216 - LdsNumElementsAlignedA: 38400 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 38400 - LdsOffsetB_Blk: 169472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73216 - LdsOffsetMetadata_Blk: 169472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [15, 4] - MIWaveTileA: 15 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 240 - MacroTile1: 256 - MacroTileA: 240 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 240 - NumGlobalWriteVectorsPerThread: 240 - NumLoadsA: 15 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 15 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 60 - ThreadTile1: 4 - ThreadTileA: 60 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIjK835T0h-3Ar1N2b4kmmt65KtYLTJhfaZH3LmR2auFA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 68096 - LdsInitCVgprs: false - LdsNumBytes: 68096 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 68096 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [13, 4] - MIWaveTileA: 13 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 208 - MacroTile1: 256 - MacroTileA: 208 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 208 - NumLoadsA: 13 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 13 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 52 - ThreadTile1: 4 - ThreadTileA: 52 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIgsJynaoGp5Lr3OC9AUIiNlgAW9kkq7qPUmgK4ohOA_k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 128512 - LdsInitCVgprs: false - LdsNumBytes: 128512 - LdsNumElementsAlignedA: 28160 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 28160 - LdsOffsetB_Blk: 93696 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 28160 - LdsOffsetMetadata_Blk: 93696 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [11, 4] - MIWaveTileA: 11 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 176 - MacroTile1: 256 - MacroTileA: 176 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 176 - NumLoadsA: 11 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 11 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 44 - ThreadTile1: 4 - ThreadTileA: 44 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI6ZUiH-qsm4LTRuwyb7uw6oTXUaPI3Vjpt7WHaUbBOfI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123392 - LdsInitCVgprs: false - LdsNumBytes: 123392 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 88576 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23040 - LdsOffsetMetadata_Blk: 88576 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [9, 4] - MIWaveTileA: 9 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 144 - MacroTile1: 256 - MacroTileA: 144 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 144 - NumLoadsA: 9 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 9 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 36 - ThreadTile1: 4 - ThreadTileA: 36 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIiu6Dap7c-mOezayg2ePWEWJWIUNWwq_GUAWuc6asW8k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 17920 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17920 - LdsOffsetB_Blk: 83456 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 83456 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 112 - MacroTile1: 256 - MacroTileA: 112 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1fyJI8fnAEv3sOPHcVPqOC2AU9kSqOftCPxastr6HZJE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 47616 - LdsInitCVgprs: false - LdsNumBytes: 47616 - LdsNumElementsAlignedA: 12800 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12800 - LdsOffsetB_Blk: 78336 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 47616 - LdsOffsetMetadata_Blk: 78336 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-Zo3r6x3T98ZMWTZ4H7LqLtk7xXmH_6zOjhPN09FdTk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 7680 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 7680 - LdsOffsetB_Blk: 73216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 73216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1wehOWUghk2A68IMMHO37hmuhc-kFrA6X9yfb9WRqhBg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37376 - LdsInitCVgprs: false - LdsNumBytes: 37376 - LdsNumElementsAlignedA: 2560 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2560 - LdsOffsetB_Blk: 68096 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37376 - LdsOffsetMetadata_Blk: 68096 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI1vCzuwEprjPexlyglCW3zRXsjX9xUmGeZzokhH60Z9yE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 75776 - LdsInitCVgprs: false - LdsNumBytes: 75776 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 75776 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI1SiLGOArtDjYGdvuIvIxglRBR2PASF7znHipyq968J9k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81920 - LdsInitCVgprs: false - LdsNumBytes: 81920 - LdsNumElementsAlignedA: 64512 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 64512 - LdsOffsetB_Blk: 195584 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81920 - LdsOffsetMetadata_Blk: 195584 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 14 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 14 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI1Atkrqghs5hZ7rFrRDMn1qRHk9P7RjRIX2jTHJaCZcIs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 64512 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 64512 - LdsOffsetB_Blk: 195584 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 195584 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 14 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 14 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI1H6kBMzYhDsiXY1vBEU2Ud8UhLYehCYuWeQaCulrUTHo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 12 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI153vNbJPjGrVniCPDOzS5UGzSK6NOLhTT3dYrLoo1YJ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 12 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1y0BaulKQmTtKr2XSp5P3dnaOK0k5UYUxfkhT1HxC_rU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 117760 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 117760 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MIUYoy0JGLYkrhIavz9gjog-Elk__Z0jYq-QXQEcN2PaA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 177152 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 177152 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 10 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI1cUVPZBoZ-mtmE8sTWF_Az_FU-gB_6I8zhwfiN1pEaks= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 177152 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 177152 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 10 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI1EM3GcfaRjwnLLnXwjtnf8XZHB699buVut-TXk4ZjWyc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 63488 - LdsInitCVgprs: false - LdsNumBytes: 63488 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 63488 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 10 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 88 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI15N7cdBjXALKtpaYkT17ouFsMqBdqYm1rqUc2cC7GeLE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 10 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 89 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MIU14cbwZ_t9NC-mBYr7N5MljJmeoGwBb-WUmjyy7ILhU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 90 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIM08mmLtqT14x3TxGefabTcO290EciaeH6GnjDeU8Bgo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 91 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1OG3k0SIeaxxBYmDDXCQJ0v824MXDB2cvmAH14Io8mlM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 92 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1Ee_Jq2fBrVEHXaPp-qikzlaFXpjSm2wKDPvD-MKxLM0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 93 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI13IeiqjCNSTMel5JBLjnHPiB9PE_JYWIIfXqePFminMc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 94 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI1qUn-crPudL6dXZ6rCcKnkE9s4w-RcuNNqOhYdbIrR28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 158720 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 158720 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 6 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 95 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI18wp9zZc9oencMlWPaDY7BUc_iQgMxwF-RQOzHet7J3I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 158720 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 158720 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 96 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI1LrtbILHycv46qRt6NxJYDFm4zT_v-o9V1TAKrHNeAeQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 97 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI164Fx4NhdfXwLWoN6X-Afen5a5uy0a8LeKZgpjvani3CM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 98 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16_Nc2lY4fwUC-cbMivJ1U_eNf_9vhMW0WYu6jNMKBzYI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 99 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16X2E4Wx9NVLCyWp4uy3bGT7QFxR20eVc2o2HQJqwi8Hk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 100 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI1ehl3l0UKCFUW0CqlYWmD5Gv6DpebCzV_7KPnbR_5y4Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81920 - LdsInitCVgprs: false - LdsNumBytes: 81920 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 64512 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81920 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 14 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 14 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 101 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI1PpwSxkO--tNKZnLh5u1Zqztz5GRs2qwDUtmWoQ32_0M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 102 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1BIcxM2OYKG0Jqn5GRpKyHvCNhj2aQceoHMvTm7lJo6Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 63488 - LdsInitCVgprs: false - LdsNumBytes: 63488 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 63488 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 103 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1-eL7HedP-uoMH-4uLF-vatbGDwtmp3GZhw8IuLzDHvE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 104 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16Tidk1Ou4UA72T_vnHnWAsymx18z-7WG5S7yV6EMzxAE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 105 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16P7nrnDgCqsPXJVFvUTHGSoy-i5STapWZDg4F8KHN2a8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 106 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI164FNCErQfeGGweRYwOlRLTlQQw-uihsMini7VG5MjqUs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 107 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI1Weg33rA_lR82ctIL5Wv-agG74LvDUGtagD8ZNXCm6m8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 75776 - LdsInitCVgprs: false - LdsNumBytes: 75776 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 140288 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 75776 - LdsOffsetMetadata_Blk: 140288 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 108 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI1yjWOfHPT7EqX9rhhXejY8_eiJ07brJZxCVBlCXR9DBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 64512 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 140288 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 140288 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 14 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 14 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 109 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1pVP6pjfyPWPVeerMg8FY1EyiGjH4B_j7tc4lRUzRPvg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 110 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI1q7mTED5UvVJrOPDvYFNPflDjUkX1okY5IJT3AZBddMM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 111 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1fFpYWYjPWsmrBYgoz9CIhz7OHVoysPvt9SgTel6GPWI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 112 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16mu07e3PYZ_jp2LQxX0lmvu4pmsw_265fxGpIWycVSjg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 113 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI162x9lNzvJGF2jhOPNNRp1Q3mFaUkUkcNWlM7ecLmXXfs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 114 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16-01AyvDsXfpTy-wvoK3K25-ZRKuFeCLIRsj5lVXzX_s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 18432 - LdsInitCVgprs: false - LdsNumBytes: 18432 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 18432 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 115 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI1LDs52gy7NI4WuOMQ_idbNXBwBK86orsq1YVyOo4IvKs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 90624 - LdsInitCVgprs: false - LdsNumBytes: 90624 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 90624 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 16 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 116 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1NztuuZzeaShCMYTqz64dRhvSvIEetpxjBMHS12EujLs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81408 - LdsInitCVgprs: false - LdsNumBytes: 81408 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 13824 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81408 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 16 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 117 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1-eS3GREuZbaF_WcdTrJ8tJN1VRBcjitHDgD1jEhWCcE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 72192 - LdsInitCVgprs: false - LdsNumBytes: 72192 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 4608 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 72192 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 118 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI1ZLUtUYYiOHV_3Bx1CweJ56xkOj29nbS2TUKPiIsgWeM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 90624 - LdsInitCVgprs: false - LdsNumBytes: 90624 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 154112 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 90624 - LdsOffsetMetadata_Blk: 154112 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 119 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1QlOH2GDodFrn6A0rxCKHrFBaU_QWfv580nm9xWdZY0k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81408 - LdsInitCVgprs: false - LdsNumBytes: 81408 - LdsNumElementsAlignedA: 13824 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 13824 - LdsOffsetB_Blk: 144896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81408 - LdsOffsetMetadata_Blk: 144896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 120 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1PZ1TyY_JO23lBPKX9VhACTOX37NNgYiBjf2A-tanW7w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 72192 - LdsInitCVgprs: false - LdsNumBytes: 72192 - LdsNumElementsAlignedA: 4608 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4608 - LdsOffsetB_Blk: 135680 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 72192 - LdsOffsetMetadata_Blk: 135680 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 121 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16Z6LYIKVcBKQJUXMfCTkH2UmyhUWm90vrVgFH5F5h_is= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 122 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16Gx0c38MGhJUAjADEf6sgUUCYEnSxT0HvEevn-TLzDBA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 123 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 124 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16tO84-u8CmDozhzOYFPjZx1aAnLl_RdoGaIlGdosGUzM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 125 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 126 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16nxXVLq-_LhX2N8-ggvlvGFb2CSQPdL0Jk31-3Yn8K8w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 127 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI1Uh3YqlpEBhxKyjcdbTJSpHtEarJLrJfEIa0Mnxwsb5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 8 - LoopUnroll: 1024 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 128 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16t0-mPljiEZtupRhWAFpnxBDrJen-FsvDD34Q7PN3s_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 129 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16CAJPr2P6olRKJqA2Fmiymyoh6JhX7nYYGS7ffsGJvmw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 130 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16K_Na50sAdM91rhXevAlyiM9lrmobRF_PSYV6N76rQzk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 131 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16dVDH5F1pzH8NV-uymOAiVxprnNui5yf2Mp5t1_LqLTs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 132 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16Uv-d7icfFxrZ-YUiRaTBsX3xXbc3HcmSMlEdtwh8Yb0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 133 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1TNTLRLqQ0uhz1zq5XPt3rxttGE1MT5QoFsMWFmpwZGg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 134 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI13HHr0kFgQAqJ2TjtZ8tTohe6Udz-5u0CwkMq-_9renQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 135 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1KAql7WR7THsIAKFHzmlcHQVpFofe0Vzt2xFE0wiTOfU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84480 - LdsInitCVgprs: false - LdsNumBytes: 84480 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84480 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 136 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1-LGVgPAQHxTiHAOxIZA3A-LLqgvwFhsRG4LoerQvYqg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 137 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1uwD3q8H7ROxJTYYpP8cvtJiz3DeoGvbO5IiJaZpCFnE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84480 - LdsInitCVgprs: false - LdsNumBytes: 84480 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 147968 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84480 - LdsOffsetMetadata_Blk: 147968 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 4 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 138 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MIrZs5mdCNmpikVNpV2YQQNFBCAdivVS8D1AsBXrhUP4k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 8] + MIWaveTileA: 8 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 256 + MacroTileA: 256 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 256 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 8 + ThreadTileA: 32 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MIADOZHpI0e4kz3PekZsq-nbyBZU7EZIxEg7KcvUC6g3g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 7] + MIWaveTileA: 8 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 224 + MacroTileA: 256 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 224 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 7 + ThreadTileA: 32 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIszVB7i9sCthoswarzra0qLsMXdGUHQA9HKGzu-1z5Kk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 6] + MIWaveTileA: 8 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 192 + MacroTileA: 256 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 6 + ThreadTileA: 32 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI4oZDINgtVU-f19tz2HYWh_LmdGFiZplMuFYBiLKIRVE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 124928 + LdsInitCVgprs: false + LdsNumBytes: 124928 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 5] + MIWaveTileA: 8 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 160 + MacroTileA: 256 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 5 + ThreadTileA: 32 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIQveu_FRPQ7kB2acKv85tmm0lWhIAYF1e-bVgSb8HhRU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1nNTR_ZcKmsj4A34uOMZTZX_oilYaQuaQFL2SNhs8cXo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 49152 + LdsInitCVgprs: false + LdsNumBytes: 49152 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 49152 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI17E3Elp9R2iGRIjoSUJt7ENU0IkIEDCYD27G5goO_zfQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI14oL9i2wAQa7K6yEREmecBQB_z40_9_5z5qamIYTi9BA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 38912 + LdsInitCVgprs: false + LdsNumBytes: 38912 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38912 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MIVqWf-zHb4fGrqaI_PNfaMIuBSbf24WaOtDVKEmRwTy0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 166912 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 166912 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 8] + MIWaveTileA: 7 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 256 + MacroTileA: 224 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 224 + NumGlobalWriteVectorsPerThread: 224 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 8 + ThreadTileA: 28 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI2FE1_xdi6gisBeCfrvgksfzzbx9Yx6-LxacJqFoorOs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 71680 + LdsInitCVgprs: false + LdsNumBytes: 71680 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 166912 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71680 + LdsOffsetMetadata_Blk: 166912 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 7] + MIWaveTileA: 7 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 224 + MacroTileA: 224 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 196 + NumGlobalWriteVectorsPerThread: 196 + NumLoadsA: 7 + NumLoadsB: 7 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 7 + ThreadTileA: 28 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MItgCKUscKBBiMtBOmdwUt6qAHeM_MR7YbKWEkh39p8UQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 129024 + LdsInitCVgprs: false + LdsNumBytes: 129024 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 6] + MIWaveTileA: 7 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 192 + MacroTileA: 224 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 168 + NumGlobalWriteVectorsPerThread: 168 + NumLoadsA: 7 + NumLoadsB: 6 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 6 + ThreadTileA: 28 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI_UOU6HZrGnyaniPidkPyA2vc3eMgOLn6XHhdMyr3gok= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 5] + MIWaveTileA: 7 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 160 + MacroTileA: 224 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 140 + NumGlobalWriteVectorsPerThread: 140 + NumLoadsA: 7 + NumLoadsB: 5 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 5 + ThreadTileA: 28 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIdhuZzZrmslv6fIdikHYnhn1dpXNXnkfYYXLJoFZzt6s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 128 + MacroTileA: 224 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 4 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1b5RoEaxogDkty9pH0a4h4iqEK74VHsxBERA-H-NeUNY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 3] + MIWaveTileA: 7 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 96 + MacroTileA: 224 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 7 + NumLoadsB: 3 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 3 + ThreadTileA: 28 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1g3lxPhwHBbhj-5_vMz-ezv_Gct2IZefQGw3UsfUuGuM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 7 + NumLoadsB: 2 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1pRHgvlAR7N2hth7GOj12UHkPW1O97Tnl14UAnpJSv9c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 7 + NumLoadsB: 1 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICt7le9q16ngZmPo_ZNYkoWOjOg2jVjonbKgLeTEzLxQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 8] + MIWaveTileA: 6 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 256 + MacroTileA: 192 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 8 + ThreadTileA: 24 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MIFCEugnbAE1o_KNOxQTjmgk1-riRp7-bqr1XYLHQkDJI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 129024 + LdsInitCVgprs: false + LdsNumBytes: 129024 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 7] + MIWaveTileA: 6 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 224 + MacroTileA: 192 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 168 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 6 + NumLoadsB: 7 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 7 + ThreadTileA: 24 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MIsXAzSnuik23-u_Tqj9IA9yIVlTTHvUnKVAh5tdPfE-0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 120832 + LdsInitCVgprs: false + LdsNumBytes: 120832 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 6] + MIWaveTileA: 6 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 192 + MacroTileA: 192 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 6 + NumLoadsB: 6 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 6 + ThreadTileA: 24 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIdawCsxOV8Q1yfZkDghTj5ROObRZEHHk5r8TLSJ8-4DI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 5] + MIWaveTileA: 6 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 160 + MacroTileA: 192 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 5 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 5 + ThreadTileA: 24 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI96PVzcjbGKsmGAPHOizb5QovZZPMnBXqYWXYWCMR6RE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1XF9bGkyS2smPrKpZKRvXqZE_IlG80gFUg26F5mTZLis= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1cyGRGs5fNOjzNULtz2NEdDFcpvgD-53w5wck9Mz0rEA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1dXOVrMlvBwO6y7ESILkUB5p9GtB-Km9ykpVzg9YqhsY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 60416 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 60416 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI_FMKfIaIlw04mnTidFqT-Fnj0gDbPBKMLBy9AxGoipc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 124928 + LdsInitCVgprs: false + LdsNumBytes: 124928 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 8] + MIWaveTileA: 5 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 256 + MacroTileA: 160 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 160 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 8 + ThreadTileA: 20 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MIgvcBHckHBXx7rYEXvd58PMUeCwZSGsDJtMgKSLGOngQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 7] + MIWaveTileA: 5 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 224 + MacroTileA: 160 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 140 + NumGlobalWriteVectorsPerThread: 140 + NumLoadsA: 5 + NumLoadsB: 7 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 7 + ThreadTileA: 20 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIiC0cBFJQlkymU09JdEq7dUq1wvcSX9wghKOSQxkOErI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 6] + MIWaveTileA: 5 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 192 + MacroTileA: 160 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 120 + NumLoadsA: 5 + NumLoadsB: 6 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 6 + ThreadTileA: 20 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MIwREMu9dXpHu0hNi1Te8Ryumj0HYpn1BFkusQoNfhAQ8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 5] + MIWaveTileA: 5 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 160 + MacroTileA: 160 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 100 + NumGlobalWriteVectorsPerThread: 100 + NumLoadsA: 5 + NumLoadsB: 5 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 5 + ThreadTileA: 20 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIcrglZoxFbHHjwhvMHFjwAoFdYaccIqWMvEsbqy0yalI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1BUsuevlFp3qE9qBkCpl8ZHtyZp2nTMvZX1nz1cQRT7M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1IVM_kq6fQOztyL87pw0KKbmpiglUkaPXmejIGutUrh4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1bgH1ukcWIV-QEjZUrEMgbOOF-CIy8Yu_mpQb3L9Zq3g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 58368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 58368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU8_SUM1_SUS256_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM8_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 51712 + LdsInitCVgprs: false + LdsNumBytes: 51712 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51712 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [32, 32, 64, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 64 + MatrixInstM: 32 + MatrixInstN: 32 + MatrixInstruction: [32, 32, 64, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NonTemporalC: 4 + NonTemporalD: 4 + NumElementsPerBatchStore: 16 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 8 + StaggerUMapping: 1 + StorePriorityOpt: true + StoreSwapAddr: false + StoreSyncOpt: 4 + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 2 + SubGroup1: 128 + SubGroupA: 2 + SubGroupB: 128 + ThreadTile: [1, 1] + ThreadTile0: 64 + ThreadTile1: 2 + ThreadTileA: 64 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 1 + enableLDSTrA: 0 + enableLDSTrB: 0 + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIHAxeqAl5GZyYpWyj9YBlHLiAot0B1CVx9Nf_5OHRC1Q= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 224 + MacroTileA: 128 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 7 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIIlrnP4n1WnSi3fe0xoxepRfTuuLMMp-jgbnYdGnnvYE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI5bZ7yznefmkxpzzbB-XiEKlKRJu2AMeEgvXx60Tq5L0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI72iEPgzvyM2ZrWa9mJE7BjRqJtpIrgM81I_tGNEjK0c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1xf_ukgQ-Z-qNS4jmkabhCAr3mB7WAQFBVo0jxB_s6wM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1j8uVA7i_pGx8ZCBrMR8ZhK3U6bp7J7YjtbYnnKdPxDI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1M6uR2wC2h6sr2Hl9YrF8zhWGQ7RgoVsQ-TM7lCE6pPI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 22528 + LdsInitCVgprs: false + LdsNumBytes: 22528 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22528 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1lw4esJJcKXjrN7F_gKTEASvr4FPioQKVPOZImphrvBY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 49152 + LdsInitCVgprs: false + LdsNumBytes: 49152 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 49152 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1JixbiZckG_HlXVPjjWZHNInPUwLf2xyuU7BTFa5yHI0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 7] + MIWaveTileA: 3 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 224 + MacroTileA: 96 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 3 + NumLoadsB: 7 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 7 + ThreadTileA: 12 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1wIqbeUYowBvH47eY0D90voHkVaoBg2lPdbX2d68nIJ4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1J-2M-3q4LSCntZjUh3v8nZH3qUcDC89rCJ0Fdq3iMeI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1Vr1CG6o8iRf8BbhZJxSu0FVBXoX31LtlPuM8cs_YEXU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16Cnahg_B3KmoGxZkTO1OYZLBTHOrwHNY1VQd2t5IrylI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16W4yMmcRT_Rcp3ZNo2gSW3q4SJdd9i9DRRKcgqox59VE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 24576 + LdsInitCVgprs: false + LdsNumBytes: 24576 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 24576 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16zNms4_iu6xgFQiq42WUbXrNnFNyslp04sbAKWAlTbog= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 20480 + LdsInitCVgprs: false + LdsNumBytes: 20480 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20480 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI11aFcbk23grO4-hK8aBkA0Q2TJ_RzSnSefhCsshDFWE8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1BIQ9uHrC9E8FHJ9mzReo0log_q-mFe47Gvfw9lRTxKU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 7 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1oWIDEj_bU2sIdW6WZSqKJI8Okx_mH0MwWvgKg4p0Wh0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1DfS6ucItLGFYy3q0Zz-JZd7icG_ycHgazunawsT8Pns= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1jK-MUAAqzxLFJYDxWtp_lRnOH6G_Sa3LS7gsDkzJ5ZI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16qBSTIiXAL1b2k0UyETAi0pu4GSfWwKXX5o-ybzBtncE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 24576 + LdsInitCVgprs: false + LdsNumBytes: 24576 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 24576 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16mOGETtdY_W1srxs4Mfgd04W-cV79aCR6raMD3HA6jWU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 18432 + LdsInitCVgprs: false + LdsNumBytes: 18432 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 18432 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16wIclDOa00a_BcP3sOUDpDc_qFY9RvbDE777DV1a9O7k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 25600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 9216 + LdsOffsetMetadata_Blk: 25600 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1pg1aJvOuqnmTxe_-2e9_1S4uFGoAaPhgR-zs28ui024= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 38912 + LdsInitCVgprs: false + LdsNumBytes: 38912 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38912 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1bJJ5iziycwP81Vg_rFbnRobq-WE2Gw-XuD7kvGV58LU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 1 + NumLoadsB: 7 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1c60kFRwy-kwCbpVB3Be-Tp-X0igfgGU9wMtdCOAL5xM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI13_eugo-sWMKarMTBe1WI4hmzJdiGvToykcHBbj4WXow= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1gSIhM4a60TQQmpGca3zPH_1C9YMOrBSb2Bu9JrxvD-k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 22528 + LdsInitCVgprs: false + LdsNumBytes: 22528 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22528 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16YH7g6N9l5w5imlP3Lp9mdDXwtO_ZtzNO9yUXMFVOjKQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 20480 + LdsInitCVgprs: false + LdsNumBytes: 20480 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20480 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16IFZMDGDs3-q6koIU2xIB_XUAfzhmng9Mg_46GO3Rqf4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16rLKaVqye5ZVlglTUO9ioHrZ6WB-GJ_Qs2RqP4S8ZIXE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MIyoGz0G9DWmdNz6WjS2Bq6rCsCrR_mXM4td8eMkMYEDw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 73216 + LdsInitCVgprs: false + LdsNumBytes: 73216 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 38400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 165888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73216 + LdsOffsetMetadata_Blk: 165888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 15] + MIWaveTileA: 4 + MIWaveTileB: 15 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 240 + MacroTileA: 256 + MacroTileB: 240 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 240 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 8 + NumLoadsB: 15 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 15 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 15 + ThreadTileA: 16 + ThreadTileB: 15 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIBnc7kWhy6AvjwbNc5adjziCfJiLAVAD1RzoNcS6OMZQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 68096 + LdsInitCVgprs: false + LdsNumBytes: 68096 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 165888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 68096 + LdsOffsetMetadata_Blk: 165888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 13] + MIWaveTileA: 4 + MIWaveTileB: 13 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 208 + MacroTileA: 256 + MacroTileB: 208 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 52 + NumLoadsA: 8 + NumLoadsB: 13 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 13 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 13 + ThreadTileA: 16 + ThreadTileB: 13 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIF5A-1D1ZIPjJsJJ7ttt4AlfEFPQG0riur7Zv_qXG5YI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 128512 + LdsInitCVgprs: false + LdsNumBytes: 128512 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 28160 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 11] + MIWaveTileA: 4 + MIWaveTileB: 11 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 176 + MacroTileA: 256 + MacroTileB: 176 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 44 + NumLoadsA: 8 + NumLoadsB: 11 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 11 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 11 + ThreadTileA: 16 + ThreadTileB: 11 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIyRCs1CYgk6qoLPVzZz2698DWR4wMJ7whNVJ8JwJfnZc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 123392 + LdsInitCVgprs: false + LdsNumBytes: 123392 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 9] + MIWaveTileA: 4 + MIWaveTileB: 9 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 144 + MacroTileA: 256 + MacroTileB: 144 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 8 + NumLoadsB: 9 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 9 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 9 + ThreadTileA: 16 + ThreadTileB: 9 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIaVDafPaNDXaLtu56b1YkZsQUI1aLmxb3jgdu_dv65EA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 17920 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 112 + MacroTileA: 256 + MacroTileB: 112 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1jNVYmoC6fjSjbIADTf0f9l0Cs4TKifqjd6Xql0ehllk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 47616 + LdsInitCVgprs: false + LdsNumBytes: 47616 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 12800 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 47616 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1eX_SljyDEcS5rCPMhk2caJwb3hhe7qqzmb_HRJ0i-U0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 7680 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1fPZ5I9MI2WZKscvBb8L9C1UGdH6oze92UNxsd_rmjZ0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 37376 + LdsInitCVgprs: false + LdsNumBytes: 37376 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 2560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37376 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI17whl_3k4Y1EhiqZzn7y-o-UTqlat-Sd3AsiBr6V3nc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 73216 + LdsInitCVgprs: false + LdsNumBytes: 73216 + LdsNumElementsAlignedA: 38400 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 38400 + LdsOffsetB_Blk: 169472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73216 + LdsOffsetMetadata_Blk: 169472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [15, 4] + MIWaveTileA: 15 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 240 + MacroTile1: 256 + MacroTileA: 240 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 240 + NumGlobalWriteVectorsPerThread: 240 + NumLoadsA: 15 + NumLoadsB: 8 + NumLoadsPerpendicularA: 15 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 60 + ThreadTile1: 4 + ThreadTileA: 60 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIjK835T0h-3Ar1N2b4kmmt65KtYLTJhfaZH3LmR2auFA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 68096 + LdsInitCVgprs: false + LdsNumBytes: 68096 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 68096 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [13, 4] + MIWaveTileA: 13 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 208 + MacroTile1: 256 + MacroTileA: 208 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 208 + NumLoadsA: 13 + NumLoadsB: 8 + NumLoadsPerpendicularA: 13 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 52 + ThreadTile1: 4 + ThreadTileA: 52 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIgsJynaoGp5Lr3OC9AUIiNlgAW9kkq7qPUmgK4ohOA_k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 128512 + LdsInitCVgprs: false + LdsNumBytes: 128512 + LdsNumElementsAlignedA: 28160 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 28160 + LdsOffsetB_Blk: 93696 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 28160 + LdsOffsetMetadata_Blk: 93696 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [11, 4] + MIWaveTileA: 11 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 176 + MacroTile1: 256 + MacroTileA: 176 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 176 + NumLoadsA: 11 + NumLoadsB: 8 + NumLoadsPerpendicularA: 11 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 44 + ThreadTile1: 4 + ThreadTileA: 44 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI6ZUiH-qsm4LTRuwyb7uw6oTXUaPI3Vjpt7WHaUbBOfI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 123392 + LdsInitCVgprs: false + LdsNumBytes: 123392 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 88576 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23040 + LdsOffsetMetadata_Blk: 88576 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [9, 4] + MIWaveTileA: 9 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 144 + MacroTile1: 256 + MacroTileA: 144 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 144 + NumLoadsA: 9 + NumLoadsB: 8 + NumLoadsPerpendicularA: 9 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 36 + ThreadTile1: 4 + ThreadTileA: 36 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIiu6Dap7c-mOezayg2ePWEWJWIUNWwq_GUAWuc6asW8k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 17920 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17920 + LdsOffsetB_Blk: 83456 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 83456 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 112 + MacroTile1: 256 + MacroTileA: 112 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1fyJI8fnAEv3sOPHcVPqOC2AU9kSqOftCPxastr6HZJE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 47616 + LdsInitCVgprs: false + LdsNumBytes: 47616 + LdsNumElementsAlignedA: 12800 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12800 + LdsOffsetB_Blk: 78336 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 47616 + LdsOffsetMetadata_Blk: 78336 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-Zo3r6x3T98ZMWTZ4H7LqLtk7xXmH_6zOjhPN09FdTk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 7680 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 7680 + LdsOffsetB_Blk: 73216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 73216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1wehOWUghk2A68IMMHO37hmuhc-kFrA6X9yfb9WRqhBg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37376 + LdsInitCVgprs: false + LdsNumBytes: 37376 + LdsNumElementsAlignedA: 2560 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2560 + LdsOffsetB_Blk: 68096 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37376 + LdsOffsetMetadata_Blk: 68096 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI1vCzuwEprjPexlyglCW3zRXsjX9xUmGeZzokhH60Z9yE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 75776 + LdsInitCVgprs: false + LdsNumBytes: 75776 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 75776 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 2 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI1SiLGOArtDjYGdvuIvIxglRBR2PASF7znHipyq968J9k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 81920 + LdsInitCVgprs: false + LdsNumBytes: 81920 + LdsNumElementsAlignedA: 64512 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 64512 + LdsOffsetB_Blk: 195584 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81920 + LdsOffsetMetadata_Blk: 195584 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 14 + NumLoadsB: 4 + NumLoadsPerpendicularA: 14 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI1Atkrqghs5hZ7rFrRDMn1qRHk9P7RjRIX2jTHJaCZcIs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 64512 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 64512 + LdsOffsetB_Blk: 195584 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 195584 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 14 + NumLoadsB: 2 + NumLoadsPerpendicularA: 14 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI1H6kBMzYhDsiXY1vBEU2Ud8UhLYehCYuWeQaCulrUTHo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 12 + NumLoadsB: 6 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI153vNbJPjGrVniCPDOzS5UGzSK6NOLhTT3dYrLoo1YJ4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 12 + NumLoadsB: 4 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1y0BaulKQmTtKr2XSp5P3dnaOK0k5UYUxfkhT1HxC_rU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 117760 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 117760 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 2 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MIUYoy0JGLYkrhIavz9gjog-Elk__Z0jYq-QXQEcN2PaA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 177152 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 177152 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 10 + NumLoadsB: 8 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI1cUVPZBoZ-mtmE8sTWF_Az_FU-gB_6I8zhwfiN1pEaks= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 177152 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 177152 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 10 + NumLoadsB: 6 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 88 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI1EM3GcfaRjwnLLnXwjtnf8XZHB699buVut-TXk4ZjWyc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 63488 + LdsInitCVgprs: false + LdsNumBytes: 63488 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 63488 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 10 + NumLoadsB: 4 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 89 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI15N7cdBjXALKtpaYkT17ouFsMqBdqYm1rqUc2cC7GeLE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 10 + NumLoadsB: 2 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 90 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MIU14cbwZ_t9NC-mBYr7N5MljJmeoGwBb-WUmjyy7ILhU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 10 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 91 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIM08mmLtqT14x3TxGefabTcO290EciaeH6GnjDeU8Bgo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 92 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1OG3k0SIeaxxBYmDDXCQJ0v824MXDB2cvmAH14Io8mlM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 93 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1Ee_Jq2fBrVEHXaPp-qikzlaFXpjSm2wKDPvD-MKxLM0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 94 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI13IeiqjCNSTMel5JBLjnHPiB9PE_JYWIIfXqePFminMc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 95 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI1qUn-crPudL6dXZ6rCcKnkE9s4w-RcuNNqOhYdbIrR28= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 158720 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 158720 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 6 + NumLoadsB: 12 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 96 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI18wp9zZc9oencMlWPaDY7BUc_iQgMxwF-RQOzHet7J3I= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 158720 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 158720 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 10 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 97 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI1LrtbILHycv46qRt6NxJYDFm4zT_v-o9V1TAKrHNeAeQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 98 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI164Fx4NhdfXwLWoN6X-Afen5a5uy0a8LeKZgpjvani3CM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 6 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 99 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16_Nc2lY4fwUC-cbMivJ1U_eNf_9vhMW0WYu6jNMKBzYI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 100 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16X2E4Wx9NVLCyWp4uy3bGT7QFxR20eVc2o2HQJqwi8Hk= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 101 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI1ehl3l0UKCFUW0CqlYWmD5Gv6DpebCzV_7KPnbR_5y4Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 81920 + LdsInitCVgprs: false + LdsNumBytes: 81920 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 64512 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81920 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 14 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 14 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 102 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI1PpwSxkO--tNKZnLh5u1Zqztz5GRs2qwDUtmWoQ32_0M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 12 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 103 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1BIcxM2OYKG0Jqn5GRpKyHvCNhj2aQceoHMvTm7lJo6Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 63488 + LdsInitCVgprs: false + LdsNumBytes: 63488 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 63488 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 10 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 104 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1-eL7HedP-uoMH-4uLF-vatbGDwtmp3GZhw8IuLzDHvE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 105 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16Tidk1Ou4UA72T_vnHnWAsymx18z-7WG5S7yV6EMzxAE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 106 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16P7nrnDgCqsPXJVFvUTHGSoy-i5STapWZDg4F8KHN2a8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 107 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI164FNCErQfeGGweRYwOlRLTlQQw-uihsMini7VG5MjqUs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 108 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI1Weg33rA_lR82ctIL5Wv-agG74LvDUGtagD8ZNXCm6m8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 75776 + LdsInitCVgprs: false + LdsNumBytes: 75776 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 140288 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 75776 + LdsOffsetMetadata_Blk: 140288 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 16 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 109 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI1yjWOfHPT7EqX9rhhXejY8_eiJ07brJZxCVBlCXR9DBo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 64512 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 140288 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 140288 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 14 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 14 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 110 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1pVP6pjfyPWPVeerMg8FY1EyiGjH4B_j7tc4lRUzRPvg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 12 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 111 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI1q7mTED5UvVJrOPDvYFNPflDjUkX1okY5IJT3AZBddMM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 10 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 112 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1fFpYWYjPWsmrBYgoz9CIhz7OHVoysPvt9SgTel6GPWI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 113 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16mu07e3PYZ_jp2LQxX0lmvu4pmsw_265fxGpIWycVSjg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 114 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI162x9lNzvJGF2jhOPNNRp1Q3mFaUkUkcNWlM7ecLmXXfs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 115 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16-01AyvDsXfpTy-wvoK3K25-ZRKuFeCLIRsj5lVXzX_s= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 18432 + LdsInitCVgprs: false + LdsNumBytes: 18432 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 18432 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 116 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI1LDs52gy7NI4WuOMQ_idbNXBwBK86orsq1YVyOo4IvKs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 90624 + LdsInitCVgprs: false + LdsNumBytes: 90624 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 90624 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 16 + NumLoadsB: 5 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 117 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1NztuuZzeaShCMYTqz64dRhvSvIEetpxjBMHS12EujLs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 81408 + LdsInitCVgprs: false + LdsNumBytes: 81408 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 13824 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81408 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 16 + NumLoadsB: 3 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 118 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1-eS3GREuZbaF_WcdTrJ8tJN1VRBcjitHDgD1jEhWCcE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 72192 + LdsInitCVgprs: false + LdsNumBytes: 72192 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 4608 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 72192 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 119 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI1ZLUtUYYiOHV_3Bx1CweJ56xkOj29nbS2TUKPiIsgWeM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 90624 + LdsInitCVgprs: false + LdsNumBytes: 90624 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 154112 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 90624 + LdsOffsetMetadata_Blk: 154112 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 16 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 120 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1QlOH2GDodFrn6A0rxCKHrFBaU_QWfv580nm9xWdZY0k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 81408 + LdsInitCVgprs: false + LdsNumBytes: 81408 + LdsNumElementsAlignedA: 13824 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 13824 + LdsOffsetB_Blk: 144896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81408 + LdsOffsetMetadata_Blk: 144896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 16 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 121 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1PZ1TyY_JO23lBPKX9VhACTOX37NNgYiBjf2A-tanW7w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 72192 + LdsInitCVgprs: false + LdsNumBytes: 72192 + LdsNumElementsAlignedA: 4608 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4608 + LdsOffsetB_Blk: 135680 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 72192 + LdsOffsetMetadata_Blk: 135680 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 122 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16Z6LYIKVcBKQJUXMfCTkH2UmyhUWm90vrVgFH5F5h_is= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 4 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 123 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16Gx0c38MGhJUAjADEf6sgUUCYEnSxT0HvEevn-TLzDBA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 124 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 125 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16tO84-u8CmDozhzOYFPjZx1aAnLl_RdoGaIlGdosGUzM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 12 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 126 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 127 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16nxXVLq-_LhX2N8-ggvlvGFb2CSQPdL0Jk31-3Yn8K8w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 128 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI1Uh3YqlpEBhxKyjcdbTJSpHtEarJLrJfEIa0Mnxwsb5k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 8 + LoopUnroll: 1024 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 129 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16t0-mPljiEZtupRhWAFpnxBDrJen-FsvDD34Q7PN3s_4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 130 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16CAJPr2P6olRKJqA2Fmiymyoh6JhX7nYYGS7ffsGJvmw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 131 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16K_Na50sAdM91rhXevAlyiM9lrmobRF_PSYV6N76rQzk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 132 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16dVDH5F1pzH8NV-uymOAiVxprnNui5yf2Mp5t1_LqLTs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 133 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16Uv-d7icfFxrZ-YUiRaTBsX3xXbc3HcmSMlEdtwh8Yb0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 134 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1TNTLRLqQ0uhz1zq5XPt3rxttGE1MT5QoFsMWFmpwZGg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 135 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI13HHr0kFgQAqJ2TjtZ8tTohe6Udz-5u0CwkMq-_9renQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 136 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1KAql7WR7THsIAKFHzmlcHQVpFofe0Vzt2xFE0wiTOfU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84480 + LdsInitCVgprs: false + LdsNumBytes: 84480 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84480 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 4 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 137 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1-LGVgPAQHxTiHAOxIZA3A-LLqgvwFhsRG4LoerQvYqg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 138 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM6_WGMXCC1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1uwD3q8H7ROxJTYYpP8cvtJiz3DeoGvbO5IiJaZpCFnE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84480 + LdsInitCVgprs: false + LdsNumBytes: 84480 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 147968 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84480 + LdsOffsetMetadata_Blk: 147968 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalB: 4 + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 16 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index fa4149498fd6..bcf59e47d664 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,11034 +81,7994 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1zEvSrUfniyIPA5Nficsug_rlEaVAGxHNYFMuvmYifqc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1oq7eVBWwzey19A8uIqsDkLEvAC8u4jm_3WdeHmyqNY8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1RipBYLITmgLj7nJiXhyI57h-QdswAlXmvZRQ-t7Yij4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33280 - LdsInitCVgprs: false - LdsNumBytes: 33280 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1iIhyCvIYW4EQ_x86i0IZ86zDKGdESkaJ9YYcm0hSGzI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 57600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 57600 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1eZeOxd9MrHLQR2GNA65EfarGb_PDYi0bDvnnTj3iymw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1fok8NkYtMokxqZZMiJ_iGKUDQvUnmdxY1flvKGmyHyE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI4THos_lFJzWi_Dnch9qAlPGNCCGbpGTJT_pt2NL4ZB8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1pEJy9yEKZBXA9EcPO5rjF-8vFHW5-X8UkNyH2oVn994= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29440 - LdsInitCVgprs: false - LdsNumBytes: 29440 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29440 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1S6e2ViSr3lOZh36puDEQ4ogQb2Kj5RPE7WzL1lq3F0M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1C6jTq6qNnho1ENixE4x2q_23xMvUAX0Q3bKaL_ah5XQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21248 - LdsInitCVgprs: false - LdsNumBytes: 21248 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21248 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1158FxB1URc5cpKzdBy5PKmwBlaNnYvU5pNK2aJTi3FY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29440 - LdsInitCVgprs: false - LdsNumBytes: 29440 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29440 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI163lHbClsVTGPnNSzP8QvKwmggFBp2OvBLf9IoiXlJZAY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16cyaq9MxU7sNbLEazHdXIfoficQy4YOPE4TFqMztilgY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20992 - LdsInitCVgprs: false - LdsNumBytes: 20992 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20992 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16O3R5PeDu0JjUKDdGyCc1CoE20Q-U4WqcAOA_B9UwqIo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1LPyOUj4mpprhT6Es7dZqjCKeRoX1IDrnjJ3fYZl1HPI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1dsiiVyF-YL7QpBTxukjNXBE5_N2gPbhGhfp7ioSoYBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33280 - LdsInitCVgprs: false - LdsNumBytes: 33280 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1w-MN2Mm1J0St8q8s9xFauQAWxA0khy1Pd4WoSoekSlg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1cgKWeOcB-B078ImMnsIo0ZS6zy_3JERNpCWjdkytEyw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16_rKHyj2ip3IiAlJ-oI4Mi3S3Jk5OCWj2H83MWBg-Vss= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20992 - LdsInitCVgprs: false - LdsNumBytes: 20992 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20992 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16Qa_B98mEcapAL2iX2YgOcoJisssJyHDlM9u9CPAkaUM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16EwXWjXKR9ATyJNfH3DyoYi3LQ4H0ugLmKOZi9zzkKhc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29056 - LdsInitCVgprs: false - LdsNumBytes: 29056 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 24832 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 8448 - LdsOffsetMetadata_Blk: 24832 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI15LrCqHq2DyagcJd2X5qTNiDHWQnVHm_AqMqcFDlvmsU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1RS5-E3dcQYltJiaIWglbCOB49DJT8EKGQr29y9ht3jk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29184 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1WAM7dQoYboHh0SapFH0XK75zgHRfXzHLhlFGHRFgW8U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25088 - LdsInitCVgprs: false - LdsNumBytes: 25088 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25088 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1fkykvXwBDcTCeZQcsORsfStQSOevaJqa0Sia79tVqpE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21248 - LdsInitCVgprs: false - LdsNumBytes: 21248 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21248 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16WEwxgg8Y4FSUK3HPuX6yKvRQPHQ0p3bui3ZyCjYgs5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16PYalQdoVElqz6wbixR1EPrpAfE0UNhfwtrN4ZWmHKYs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29184 - LdsInitCVgprs: false - LdsNumBytes: 29184 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16XYtEgd7zoKly5GXv2t2pAYLX-e76OQRyOB6Ygozeu0A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24960 - LdsInitCVgprs: false - LdsNumBytes: 24960 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1JWVR-jh-aLg0rrdtghL_Hme5Pul5fnhSrwkCOBuPtIU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 768 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 39680 - LdsInitCVgprs: false - LdsNumBytes: 39680 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 6400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 39680 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1PwJ42uBJtJDXBA5J9eUpFXnSBdG873yhqH9x9xHr9aM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35584 - LdsInitCVgprs: false - LdsNumBytes: 35584 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 2304 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35584 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-ckiACWa5_Hl9q0vruUn7fAZYeT9xW9mnDjkhP2c50s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 768 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 39680 - LdsInitCVgprs: false - LdsNumBytes: 39680 - LdsNumElementsAlignedA: 6400 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 6400 - LdsOffsetB_Blk: 71936 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 39680 - LdsOffsetMetadata_Blk: 71936 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1MnL3tvF_jGDAY_I8anP7ysG0uEcoQof1urIS_BrpX_k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35584 - LdsInitCVgprs: false - LdsNumBytes: 35584 - LdsNumElementsAlignedA: 2304 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2304 - LdsOffsetB_Blk: 67840 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35584 - LdsOffsetMetadata_Blk: 67840 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1arXBxIFzyAilDtP1DqyksxdPITOkluw9jmLU5u0Tt2w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1rXabrsjNwkup_JrxozMVR_3U4GdmTFKff0KArypEYAA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1VT_L439A1JYsr4mdupck3knPO5y9YC1PTbi05SL_Cz0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16zQIQS-CrV5B-hJqIIcfIrW4-99zhrrQJDstr_sFS9t4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 64 - LSPA: 64 - LSPB: 64 - LVCA: 4 - LVCB: 4 - LVPA: 4 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16yzgS8-MSN2UCxzxDcSy8jzQVwvB-olaZ82o1J7M6llU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 32 - LSPA: 64 - LSPB: 128 - LVCA: 4 - LVCB: 2 - LVPA: 4 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI12hFMSt6lMTI8oWW_7aZTwxjiukY6LB_x-MEbbxd5p3M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI163QE_QBcfCjhjm6YvZW1aBPJMqF6E2nryKoAjsugzB0U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 64 - LSPA: 128 - LSPB: 64 - LVCA: 2 - LVCB: 4 - LVPA: 8 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25344 - LdsInitCVgprs: false - LdsNumBytes: 25344 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25344 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16CqP0rVSFO3WDeB13qT6LUGERdrWprMFCekioRFqiTqQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 128 - LSPB: 128 - LVCA: 2 - LVCB: 2 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 16896 - LdsInitCVgprs: false - LdsNumBytes: 16896 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 16896 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1h8DSl_KtVz4vyL0LXehy8GQa575LmnT-GS8Sm349l8c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 256 - LVCA: 16 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 70912 - LdsInitCVgprs: false - LdsNumBytes: 70912 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 70912 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1ABd-H42vx6NZOkifL5ShxVhgON-oxRsJ0GyaVxfUhIg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 256 - LSPB: 16 - LVCA: 1 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 70912 - LdsInitCVgprs: false - LdsNumBytes: 70912 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 135424 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 70912 - LdsOffsetMetadata_Blk: 135424 - LdsPadA: 16 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI160Nuqi97WrEd5kWYgnYYkOYOWZ8I-jc1R0TyAYU3gxQk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 16 - LSPA: 256 - LSPB: 256 - LVCA: 1 - LVCB: 1 - LVPA: 16 - LVPB: 16 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16y2XC2HjmywvRNMRzvdc4xwYo7V4GZ1mS8mdECpGkL28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 16 - LSPA: 128 - LSPB: 256 - LVCA: 2 - LVCB: 1 - LVPA: 8 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI162YLFRDLjA2hwymDeM7kI4wEg1_Ytlga74L2JT4VF0MM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 64 - LSCB: 16 - LSPA: 64 - LSPB: 256 - LVCA: 4 - LVCB: 1 - LVPA: 4 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI1651AQfSOaqL9x7qKXZbPBT7ZWUW0NRJycR149B4OSQTc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 32 - LSPA: 256 - LSPB: 128 - LVCA: 1 - LVCB: 2 - LVPA: 16 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI1674YMvMGkXuXL1Cf1gzLE5KEGPC_Zl12HLRkso0LE7_M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 64 - LSPA: 256 - LSPB: 64 - LVCA: 1 - LVCB: 4 - LVPA: 16 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1vOk7fZIE6d2Crn4j8R3f-gr9gYb6d1i7QRitxgT49e8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 16 - LSPA: 256 - LSPB: 256 - LVCA: 1 - LVCB: 1 - LVPA: 16 - LVPB: 16 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xLaYt4XA53nJMoN4GRDxGCoYkMto7pvV9s9Hmk9M424A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 0 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: 0 - UnrollMajorLDSB: 0 - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB1024_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1zEvSrUfniyIPA5Nficsug_rlEaVAGxHNYFMuvmYifqc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1oq7eVBWwzey19A8uIqsDkLEvAC8u4jm_3WdeHmyqNY8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1RipBYLITmgLj7nJiXhyI57h-QdswAlXmvZRQ-t7Yij4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33280 + LdsInitCVgprs: false + LdsNumBytes: 33280 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1iIhyCvIYW4EQ_x86i0IZ86zDKGdESkaJ9YYcm0hSGzI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 57600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 57600 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1eZeOxd9MrHLQR2GNA65EfarGb_PDYi0bDvnnTj3iymw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1fok8NkYtMokxqZZMiJ_iGKUDQvUnmdxY1flvKGmyHyE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB2048_LBSPPM0_LPA64_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI4THos_lFJzWi_Dnch9qAlPGNCCGbpGTJT_pt2NL4ZB8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1536_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1pEJy9yEKZBXA9EcPO5rjF-8vFHW5-X8UkNyH2oVn994= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 29440 + LdsInitCVgprs: false + LdsNumBytes: 29440 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29440 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1S6e2ViSr3lOZh36puDEQ4ogQb2Kj5RPE7WzL1lq3F0M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1C6jTq6qNnho1ENixE4x2q_23xMvUAX0Q3bKaL_ah5XQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21248 + LdsInitCVgprs: false + LdsNumBytes: 21248 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21248 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1158FxB1URc5cpKzdBy5PKmwBlaNnYvU5pNK2aJTi3FY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 29440 + LdsInitCVgprs: false + LdsNumBytes: 29440 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29440 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI163lHbClsVTGPnNSzP8QvKwmggFBp2OvBLf9IoiXlJZAY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16cyaq9MxU7sNbLEazHdXIfoficQy4YOPE4TFqMztilgY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 20992 + LdsInitCVgprs: false + LdsNumBytes: 20992 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20992 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16O3R5PeDu0JjUKDdGyCc1CoE20Q-U4WqcAOA_B9UwqIo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1LPyOUj4mpprhT6Es7dZqjCKeRoX1IDrnjJ3fYZl1HPI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1dsiiVyF-YL7QpBTxukjNXBE5_N2gPbhGhfp7ioSoYBo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 33280 + LdsInitCVgprs: false + LdsNumBytes: 33280 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1w-MN2Mm1J0St8q8s9xFauQAWxA0khy1Pd4WoSoekSlg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1cgKWeOcB-B078ImMnsIo0ZS6zy_3JERNpCWjdkytEyw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16_rKHyj2ip3IiAlJ-oI4Mi3S3Jk5OCWj2H83MWBg-Vss= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 20992 + LdsInitCVgprs: false + LdsNumBytes: 20992 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20992 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16Qa_B98mEcapAL2iX2YgOcoJisssJyHDlM9u9CPAkaUM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16EwXWjXKR9ATyJNfH3DyoYi3LQ4H0ugLmKOZi9zzkKhc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29056 + LdsInitCVgprs: false + LdsNumBytes: 29056 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 24832 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 8448 + LdsOffsetMetadata_Blk: 24832 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA16_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI15LrCqHq2DyagcJd2X5qTNiDHWQnVHm_AqMqcFDlvmsU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1RS5-E3dcQYltJiaIWglbCOB49DJT8EKGQr29y9ht3jk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29184 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1WAM7dQoYboHh0SapFH0XK75zgHRfXzHLhlFGHRFgW8U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 25088 + LdsInitCVgprs: false + LdsNumBytes: 25088 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25088 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1fkykvXwBDcTCeZQcsORsfStQSOevaJqa0Sia79tVqpE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 21248 + LdsInitCVgprs: false + LdsNumBytes: 21248 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21248 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16WEwxgg8Y4FSUK3HPuX6yKvRQPHQ0p3bui3ZyCjYgs5k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16PYalQdoVElqz6wbixR1EPrpAfE0UNhfwtrN4ZWmHKYs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29184 + LdsInitCVgprs: false + LdsNumBytes: 29184 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16XYtEgd7zoKly5GXv2t2pAYLX-e76OQRyOB6Ygozeu0A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 24960 + LdsInitCVgprs: false + LdsNumBytes: 24960 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB768_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1JWVR-jh-aLg0rrdtghL_Hme5Pul5fnhSrwkCOBuPtIU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 768 + LdsBytesNoAmax: 39680 + LdsInitCVgprs: false + LdsNumBytes: 39680 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 6400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 39680 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1PwJ42uBJtJDXBA5J9eUpFXnSBdG873yhqH9x9xHr9aM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 35584 + LdsInitCVgprs: false + LdsNumBytes: 35584 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 2304 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35584 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-ckiACWa5_Hl9q0vruUn7fAZYeT9xW9mnDjkhP2c50s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 768 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 39680 + LdsInitCVgprs: false + LdsNumBytes: 39680 + LdsNumElementsAlignedA: 6400 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 6400 + LdsOffsetB_Blk: 71936 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 39680 + LdsOffsetMetadata_Blk: 71936 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1MnL3tvF_jGDAY_I8anP7ysG0uEcoQof1urIS_BrpX_k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 35584 + LdsInitCVgprs: false + LdsNumBytes: 35584 + LdsNumElementsAlignedA: 2304 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2304 + LdsOffsetB_Blk: 67840 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35584 + LdsOffsetMetadata_Blk: 67840 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1arXBxIFzyAilDtP1DqyksxdPITOkluw9jmLU5u0Tt2w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1rXabrsjNwkup_JrxozMVR_3U4GdmTFKff0KArypEYAA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1VT_L439A1JYsr4mdupck3knPO5y9YC1PTbi05SL_Cz0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16zQIQS-CrV5B-hJqIIcfIrW4-99zhrrQJDstr_sFS9t4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 64 + LSPA: 64 + LSPB: 64 + LVCA: 4 + LVCB: 4 + LVPA: 4 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16yzgS8-MSN2UCxzxDcSy8jzQVwvB-olaZ82o1J7M6llU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 32 + LSPA: 64 + LSPB: 128 + LVCA: 4 + LVCB: 2 + LVPA: 4 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI12hFMSt6lMTI8oWW_7aZTwxjiukY6LB_x-MEbbxd5p3M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI163QE_QBcfCjhjm6YvZW1aBPJMqF6E2nryKoAjsugzB0U= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 64 + LSPA: 128 + LSPB: 64 + LVCA: 2 + LVCB: 4 + LVPA: 8 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25344 + LdsInitCVgprs: false + LdsNumBytes: 25344 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25344 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16CqP0rVSFO3WDeB13qT6LUGERdrWprMFCekioRFqiTqQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 128 + LSPB: 128 + LVCA: 2 + LVCB: 2 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 16896 + LdsInitCVgprs: false + LdsNumBytes: 16896 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 16896 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1h8DSl_KtVz4vyL0LXehy8GQa575LmnT-GS8Sm349l8c= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 256 + LVCA: 16 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 70912 + LdsInitCVgprs: false + LdsNumBytes: 70912 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 70912 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA16_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1ABd-H42vx6NZOkifL5ShxVhgON-oxRsJ0GyaVxfUhIg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 256 + LSPB: 16 + LVCA: 1 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 70912 + LdsInitCVgprs: false + LdsNumBytes: 70912 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 135424 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 70912 + LdsOffsetMetadata_Blk: 135424 + LdsPadA: 16 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI160Nuqi97WrEd5kWYgnYYkOYOWZ8I-jc1R0TyAYU3gxQk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 16 + LSPA: 256 + LSPB: 256 + LVCA: 1 + LVCB: 1 + LVPA: 16 + LVPB: 16 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16y2XC2HjmywvRNMRzvdc4xwYo7V4GZ1mS8mdECpGkL28= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 16 + LSPA: 128 + LSPB: 256 + LVCA: 2 + LVCB: 1 + LVPA: 8 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI162YLFRDLjA2hwymDeM7kI4wEg1_Ytlga74L2JT4VF0MM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 16 + LSPA: 64 + LSPB: 256 + LVCA: 4 + LVCB: 1 + LVPA: 4 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI1651AQfSOaqL9x7qKXZbPBT7ZWUW0NRJycR149B4OSQTc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 32 + LSPA: 256 + LSPB: 128 + LVCA: 1 + LVCB: 2 + LVPA: 16 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI1674YMvMGkXuXL1Cf1gzLE5KEGPC_Zl12HLRkso0LE7_M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 64 + LSPA: 256 + LSPB: 64 + LVCA: 1 + LVCB: 4 + LVPA: 16 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1vOk7fZIE6d2Crn4j8R3f-gr9gYb6d1i7QRitxgT49e8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 16 + LSPA: 256 + LSPB: 256 + LVCA: 1 + LVCB: 1 + LVPA: 16 + LVPB: 16 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS0_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xLaYt4XA53nJMoN4GRDxGCoYkMto7pvV9s9Hmk9M424A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 0 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: 0 + UnrollMajorLDSB: 0 + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index 059076bd84cc..30f9be511f43 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,19809 +81,14194 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIk_RqFM_AwoflFBuOWbvk-8fTqjDZ1FddXgJjbcy-G0k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 125952 - LdsInitCVgprs: false - LdsNumBytes: 125952 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 6] - MIWaveTileA: 8 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 192 - MacroTileA: 256 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 6 - ThreadTileA: 32 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MISPq2JndWaX4OL12RUN6ulsn7QcNRohtTIZC1Edgdyws= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123904 - LdsInitCVgprs: false - LdsNumBytes: 123904 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 5] - MIWaveTileA: 8 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 160 - MacroTileA: 256 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 5 - ThreadTileA: 32 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIKVXPdd7d_qWsOE3R4Nn3WRva2YN_nzMUEkSXMgI4hvI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50176 - LdsInitCVgprs: false - LdsNumBytes: 50176 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50176 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1T7zX-LAq1W4S0p0x8V1ahqv2mXqZFvICIzJD4wjn5vU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48128 - LdsInitCVgprs: false - LdsNumBytes: 48128 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48128 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1C8Q9_g8YRB2HVCaci6ZpxWtz2F-KOjbPdcOSgM2OeEI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1wcQqKyv34c9LbqUML817LBELS8jNDkpatWFuPGnAz0I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37888 - LdsInitCVgprs: false - LdsNumBytes: 37888 - LdsNumElementsAlignedA: 32768 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 32768 - LdsOffsetB_Blk: 98304 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37888 - LdsOffsetMetadata_Blk: 98304 - LdsPadA: 0 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIghxlKUHytHxV_rsO7Y3k-hKsNgsUv14Ji4M_f4zzLUg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50432 - LdsInitCVgprs: false - LdsNumBytes: 50432 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50432 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 5] - MIWaveTileA: 6 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 160 - MacroTileA: 192 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 5 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 5 - ThreadTileA: 24 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MIJBI5cmbVufum9IvszCzB765s6mkvxNBCOv5r57HrQ5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI195MLYeeCjfNjgEj-ik95KMPTRnsuVyYIPGD8iI7U3FA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1enbINkflafBFO3_Pebg7jg-XrgUNrjPKEtJ1FG_ArW4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1AXbZC25QBPC0PuokhYA5b8WyOR4HvuGr9lZsTR0HMWk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 3072 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 57600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 57600 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIYn1IItDWJZY2hmD9-0eW5TADd3akO6yu4iNVSScCFLk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 86272 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 86272 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI15qZVVldYFEVxv6DoRCJq1o-8jbrdV19RqRcT-bcTp5g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 86272 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 86272 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1lHcr20MZlAXhwNOR6Gevj4wJaRc4B-tum0bYx49prt8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1XvrvtOqqq8hKEVOCgfEFmALGzVphawbwpkAxG7nqoi8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 2560 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 20736 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 20736 - LdsOffsetB_Blk: 53504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 53504 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIYXJbKl66tz5Yz16uScZ-3QcnIi8M-PaVMoZpl5IQJR0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 8] - MIWaveTileA: 4 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 8 - ThreadTileA: 16 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIWMlYxh0gBBwfi1wvTDHJHkSEVXIuF12-m4Tq-srJGFk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 224 - MacroTileA: 128 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIU-MW4z63wAohwJUDz0C8OcgBozd0DUJEBLSSdararIA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIZLtCnUAcZQRb2Sm7BFv7A3a2vBOCNqxyJPub1dc8Z8U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MInMSU2Gh3mMVemcGZ3WqUWu2l4NG8N5rOCZZl_tZSxr4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1VCOEzWL6LPvmD0HkYK2IQmKeQ33MFQ6q3DD1gUWXPWI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32256 - LdsInitCVgprs: false - LdsNumBytes: 32256 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32256 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1bvrvP40X4hFkSo1KwPg0VEiYDXT6P8LD2PVpHtPr4t8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1KCNveOe_jYnymu00u65Iq4hQQmFEfEECKI4hYCSw86A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22016 - LdsInitCVgprs: false - LdsNumBytes: 22016 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22016 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI11tDVdUBSEo3UIS5TLyJqHVgQw3dto3agLBzaHthkf5A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46336 - LdsInitCVgprs: false - LdsNumBytes: 46336 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46336 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1uDCZdlP32w2aUIZQgU6UFaItvoNYlwEGmbTkXc1IEtc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48384 - LdsInitCVgprs: false - LdsNumBytes: 48384 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48384 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 7] - MIWaveTileA: 3 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 224 - MacroTileA: 96 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 3 - NumLoadsB: 7 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 7 - ThreadTileA: 12 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1tIdq_8HHc1AUHlfryN0_49Q8EAD9oEnK2lMgmhIKC2E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1o_9BqNra6zh_HmLvWFNQm4WYzN2W1bygUabZEe9Lzsg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 78080 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 78080 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1KePfYaceZ29tiGigvOtcCFJNsP7pueMsjzyHKVb7w64= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16tOU8-XLYrNPdvvuRaPZblzPTCq0IYpmOqRVYwuw2zIA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 27904 - LdsInitCVgprs: false - LdsNumBytes: 27904 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27904 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16iQaL89dfDqFaUwHSFuqBVL1d7gxxRqiMEISWe48rBhk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16G697WmrCkkuSW0ygicOEsjSabf2uABvGDaKBNngK4oE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 12544 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 12544 - LdsOffsetB_Blk: 45312 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 45312 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1opMCZdZFzR31a1nXMExXx-HFGwt6FHBNysNBuXPo34o= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1GPhaxX2PStJ8QfX3S-AgQFNJD6o4d_c2qqsOdt_oaC4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44288 - LdsInitCVgprs: false - LdsNumBytes: 44288 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44288 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1gqbADNhoP3bHba8oMSfqNYB-76ixvzWU8E58C_oXxBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1-VJoMMTo08ubflQERsCoO0j_TzMOoFOBMG1VV9dqzpY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI15OARqHSqH8kbGBPru-B8g6Oad-Zo3FGsUS8SxkzZZYk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16xVRyCL8pcYWjhmBX3Sksf7_ECJZKzPUO0YM7WCFd9sM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 23808 - LdsInitCVgprs: false - LdsNumBytes: 23808 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23808 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI162emMCIrLY9uAeJtwZl6HlANVDofIksu3xpgoJdG4L7I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16UNManzuffvMTmsgfWnWF6Bt5jjvtS0tBsdToDQ_bZvc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 128 - LSPA: 64 - LSPB: 32 - LVCA: 4 - LVCB: 8 - LVPA: 4 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 24832 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 8448 - LdsOffsetMetadata_Blk: 24832 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1kHn9o1wGnb0Le03rKuL4bApCYLkFlSLtoU6MjkRibzc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1NMWe8o4PG1BIiKP1ggEjZUpwkkJSDyvBBNMZ8tXiaR8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 69888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 69888 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 1 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1Q0rm-ZJ-B1Ep3tDRV7Y6v5xFppIw1pc1wB6W7qCB6eE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32000 - LdsInitCVgprs: false - LdsNumBytes: 32000 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32000 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI14_XY01zUqMSgVV3Z29zc0HvqY2RBU1XaMixpmKPTFvQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1BV068w-wQamiyACMJrCQO6AzvkFtVVaL6YUOIybq_As= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI167ogjNu2arPxWNgqgKM8sDxj_QbeHe5G3biyDU4kITPE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 19712 - LdsInitCVgprs: false - LdsNumBytes: 19712 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 37120 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 19712 - LdsOffsetMetadata_Blk: 37120 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI1606uPSrg8jSF_adB8-Q1EUFxmHwf5F9lLMz64s3UmyCk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16Jm6coo6HlXoYc23v3SkS10L-uM9lwa5bnHnY2Vs_oLo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 8 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 20736 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 4352 - LdsOffsetMetadata_Blk: 20736 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIv9D6qCD557UuGj0WYoB1KLme1Sh1mzoR8zvP-Ew1rDY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 66560 - LdsInitCVgprs: false - LdsNumBytes: 66560 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 66560 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 13] - MIWaveTileA: 4 - MIWaveTileB: 13 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 208 - MacroTileA: 256 - MacroTileB: 208 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 52 - NumLoadsA: 8 - NumLoadsB: 13 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 13 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 13 - ThreadTileA: 16 - ThreadTileB: 13 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIZTzgVMYbF8zAjkLbWPx_KguNjRPGDWzQphgfwu106mI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 28160 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 11] - MIWaveTileA: 4 - MIWaveTileB: 11 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 176 - MacroTileA: 256 - MacroTileB: 176 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 44 - NumLoadsA: 8 - NumLoadsB: 11 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 11 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 11 - ThreadTileA: 16 - ThreadTileB: 11 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIYpXgEPF2KmPhp_HSItkaLdu8ZfTql6wuzRMctpM8bWs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 121856 - LdsInitCVgprs: false - LdsNumBytes: 121856 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33280 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 9] - MIWaveTileA: 4 - MIWaveTileB: 9 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 144 - MacroTileA: 256 - MacroTileB: 144 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 8 - NumLoadsB: 9 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 9 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 9 - ThreadTileA: 16 - ThreadTileB: 9 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIM6ky9Tq6VRezat1v-Vy4Qy4OEEUGyFJHwE-XvlIpRaY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 17920 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 112 - MacroTileA: 256 - MacroTileB: 112 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1ZaT28UTzCJBM6wCUWArJk9cu4afCdA6oGhULOQC7jwQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46080 - LdsInitCVgprs: false - LdsNumBytes: 46080 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 12800 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46080 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1ckLt2U3Q4BSlwSpNR-cpd6uJ3sMr5OVPRWuKvEiSVtY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 7680 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1Cks6yoWu7mRHY5OU_L0wlyt_BkdBZ9B6KYLX0jMjEZM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35840 - LdsInitCVgprs: false - LdsNumBytes: 35840 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 2560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1LRieWm8obTv5BrugK6ZfaH9J4-zT3LrRBcx6GxwPdUk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 1280 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45312 - LdsInitCVgprs: false - LdsNumBytes: 45312 - LdsNumElementsAlignedA: 10496 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 10496 - LdsOffsetB_Blk: 76032 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45312 - LdsOffsetMetadata_Blk: 76032 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 5 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1zBVp9O_j-pDQmFJ0pruVZjPjyXjbaxIDNKuOKzE-hUk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 768 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 6400 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 6400 - LdsOffsetB_Blk: 71936 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 71936 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1LmEbjkFjRe2EHvxZ5njVJSBpnw78dqd37eC05zoq8oY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 128 - LSPA: 128 - LSPB: 32 - LVCA: 2 - LVCB: 8 - LVPA: 16 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 2304 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2304 - LdsOffsetB_Blk: 67840 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 67840 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI_Cpsqq7IdCcI6m09ixnRrSz_HTRJbwY41LkpuZ6eaow= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1tP0yLw5sOaqSqLw69T0rUISjIec4gtRcEBnLindXxd0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1WlKSkQYWc45q9-9u3Sa8KR5G5YSwdMXXvYIooja5mzU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1ZFz2Wqi9AGPfO9YZeT8hu0U-a6QRpBHZ1MNZhOuWfzs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16YTIDOkvWpHTvJMbXMK3Qn2TUcHpLDIz3L67zYqwLtxM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 1536 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 24832 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 24832 - LdsOffsetB_Blk: 90368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 90368 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 3 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1Hyiuad9E_LwogHYOQgHONBkR83eXTUE_-HwHw1z6c2M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 62976 - LdsInitCVgprs: false - LdsNumBytes: 62976 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 62976 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1_vkhGjNt0PNt_xV8VVOz2XzUCfwEcU8zljVlbKTICVg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16riuLixhIaRRQRAFH27-FM-5iTnYrSUKGh2HFdVNH4W8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16tRLsgniRf9NQ8cwUknJpPeoosk2I0euxrG_yXuRD6Zc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16zCjLhPMzebcvjlA5OUrujTBGN1OYulB3ktj1a570vLs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 256 - LSPA: 64 - LSPB: 16 - LVCA: 4 - LVCB: 16 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1OuevmgTEfRQM80_r7hzorlLPNIZiSHCEk0LhjgcOAH8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 60672 - LdsInitCVgprs: false - LdsNumBytes: 60672 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 60672 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI117HfbCeMXp_75IhCNgyVsAtudVa47Y080duX4tlb_Cc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 54528 - LdsInitCVgprs: false - LdsNumBytes: 54528 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 54528 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1w_TVfNrjE3mYvq96Bm1Jgkl-HCbXvzNQ6H6gr4PSnKc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16IX89BCLJtYzlNb-bQG51jUiesgW7gn9abBGKzaC4u7Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 73984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 73984 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16yEC0PHvmD0zrKvRDTkhqsmiXq78VvD0Eg5fcAnocoic= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16_OtB8fTrZk3Pt0bE3bXfsWfkoWXEPa2ENPfe4eaiVRI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 256 - LSPA: 128 - LSPB: 16 - LVCA: 2 - LVCB: 16 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 8448 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8448 - LdsOffsetB_Blk: 41216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41216 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1uaMQi1863lUOUemwTE99mttXxpTRkvqvmiwvtfx6Qx0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 80384 - LdsInitCVgprs: false - LdsNumBytes: 80384 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 13824 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 80384 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 16 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1_aftRJpk6u1tTs3-Hwls-Fo3A3dD889Os-hWuvfNQvw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 4096 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71168 - LdsInitCVgprs: false - LdsNumBytes: 71168 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 4608 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71168 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 64 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI18CjcSziK0PjBic2MAopWR-Ta_nHbDse6qNtds5pye7Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 16 - LSCB: 256 - LSPA: 256 - LSPB: 16 - LVCA: 1 - LVCB: 16 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71936 - LdsInitCVgprs: false - LdsNumBytes: 71936 - LdsNumElementsAlignedA: 4352 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4352 - LdsOffsetB_Blk: 135424 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71936 - LdsOffsetMetadata_Blk: 135424 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 64 - LSCB: 512 - LSPA: 64 - LSPB: 8 - LVCA: 4 - LVCB: 32 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI169OaDNKWZnXMa1R_AlPEgQpVqAFVb54TBTn3PvUMxY7A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 512 - LSPA: 128 - LSPB: 8 - LVCA: 2 - LVCB: 32 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16JRL6UviIw3ql2MWJlkjDhAPVUQuM-dfo_YP9V8INeCE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16a9rQBxUkyvDWlmJquyIGWqzMINaNpRD_aWL9QWIuzQ8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 512 - LSPA: 128 - LSPB: 8 - LVCA: 2 - LVCB: 32 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 49664 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 49664 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16Jyq1ax9kWbpfAk08jNUVf7z1XoXRVCVdpt-kBpa0fWY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 64 - LSCB: 512 - LSPA: 64 - LSPB: 8 - LVCA: 4 - LVCB: 32 - LVPA: 4 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 98816 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 98816 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16D0scrs9_FOYkVHUHqfaWIBeu8Rz72d3pOV3R8iV8H74= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI169Ao_eCjaajChoPx9i0y3MGqMyeJdbqU3dbjwB_UIKx4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 512 - LSPA: 256 - LSPB: 8 - LVCA: 1 - LVCB: 32 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1DDaFTcewwV5Ybn8swfgQtfH13aPb98dLOnRrTIJYmSQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1ifWE51xli667eSWHdf8X0Gqm1isofKMCZj08QiVl3U0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 32 - LSCB: 1024 - LSPA: 128 - LSPB: 4 - LVCA: 2 - LVCB: 64 - LVPA: 8 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1U06udU6pkFuvB0exyiBHs-wOgMdXOgbcv4nENk4uw44= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1oDw0RAqdNEFhP7idqV0O87totk9ugWGoiXKFb7JRQVw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: false - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 16 - LSCB: 1024 - LSPA: 256 - LSPB: 4 - LVCA: 1 - LVCB: 64 - LVPA: 16 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84992 - LdsInitCVgprs: false - LdsNumBytes: 84992 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84992 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 16 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xqwn4mYe9PqjtRIapRznDsflXKMnaPO3v98RcshyCRoc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 16 - LdsPadB: 8 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: false - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: false - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIk_RqFM_AwoflFBuOWbvk-8fTqjDZ1FddXgJjbcy-G0k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 125952 + LdsInitCVgprs: false + LdsNumBytes: 125952 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 6] + MIWaveTileA: 8 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 192 + MacroTileA: 256 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 6 + ThreadTileA: 32 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MISPq2JndWaX4OL12RUN6ulsn7QcNRohtTIZC1Edgdyws= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 123904 + LdsInitCVgprs: false + LdsNumBytes: 123904 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 5] + MIWaveTileA: 8 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 160 + MacroTileA: 256 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 5 + ThreadTileA: 32 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB512_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIKVXPdd7d_qWsOE3R4Nn3WRva2YN_nzMUEkSXMgI4hvI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 50176 + LdsInitCVgprs: false + LdsNumBytes: 50176 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50176 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1T7zX-LAq1W4S0p0x8V1ahqv2mXqZFvICIzJD4wjn5vU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 48128 + LdsInitCVgprs: false + LdsNumBytes: 48128 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48128 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1C8Q9_g8YRB2HVCaci6ZpxWtz2F-KOjbPdcOSgM2OeEI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA0_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1wcQqKyv34c9LbqUML817LBELS8jNDkpatWFuPGnAz0I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 37888 + LdsInitCVgprs: false + LdsNumBytes: 37888 + LdsNumElementsAlignedA: 32768 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 32768 + LdsOffsetB_Blk: 98304 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37888 + LdsOffsetMetadata_Blk: 98304 + LdsPadA: 0 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIghxlKUHytHxV_rsO7Y3k-hKsNgsUv14Ji4M_f4zzLUg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 50432 + LdsInitCVgprs: false + LdsNumBytes: 50432 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50432 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 5] + MIWaveTileA: 6 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 160 + MacroTileA: 192 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 5 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 5 + ThreadTileA: 24 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MIJBI5cmbVufum9IvszCzB765s6mkvxNBCOv5r57HrQ5k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI195MLYeeCjfNjgEj-ik95KMPTRnsuVyYIPGD8iI7U3FA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1enbINkflafBFO3_Pebg7jg-XrgUNrjPKEtJ1FG_ArW4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA3072_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1AXbZC25QBPC0PuokhYA5b8WyOR4HvuGr9lZsTR0HMWk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 3072 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 57600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 57600 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIYn1IItDWJZY2hmD9-0eW5TADd3akO6yu4iNVSScCFLk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 86272 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 86272 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI15qZVVldYFEVxv6DoRCJq1o-8jbrdV19RqRcT-bcTp5g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 86272 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 86272 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1lHcr20MZlAXhwNOR6Gevj4wJaRc4B-tum0bYx49prt8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2560_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1XvrvtOqqq8hKEVOCgfEFmALGzVphawbwpkAxG7nqoi8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 2560 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 20736 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 20736 + LdsOffsetB_Blk: 53504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 53504 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIYXJbKl66tz5Yz16uScZ-3QcnIi8M-PaVMoZpl5IQJR0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 8] + MIWaveTileA: 4 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 8 + ThreadTileA: 16 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIWMlYxh0gBBwfi1wvTDHJHkSEVXIuF12-m4Tq-srJGFk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 224 + MacroTileA: 128 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 7 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIU-MW4z63wAohwJUDz0C8OcgBozd0DUJEBLSSdararIA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIZLtCnUAcZQRb2Sm7BFv7A3a2vBOCNqxyJPub1dc8Z8U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MInMSU2Gh3mMVemcGZ3WqUWu2l4NG8N5rOCZZl_tZSxr4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1VCOEzWL6LPvmD0HkYK2IQmKeQ33MFQ6q3DD1gUWXPWI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32256 + LdsInitCVgprs: false + LdsNumBytes: 32256 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32256 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1bvrvP40X4hFkSo1KwPg0VEiYDXT6P8LD2PVpHtPr4t8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1KCNveOe_jYnymu00u65Iq4hQQmFEfEECKI4hYCSw86A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 22016 + LdsInitCVgprs: false + LdsNumBytes: 22016 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22016 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI11tDVdUBSEo3UIS5TLyJqHVgQw3dto3agLBzaHthkf5A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 46336 + LdsInitCVgprs: false + LdsNumBytes: 46336 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46336 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1uDCZdlP32w2aUIZQgU6UFaItvoNYlwEGmbTkXc1IEtc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 48384 + LdsInitCVgprs: false + LdsNumBytes: 48384 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48384 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 7] + MIWaveTileA: 3 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 224 + MacroTileA: 96 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 3 + NumLoadsB: 7 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 7 + ThreadTileA: 12 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1tIdq_8HHc1AUHlfryN0_49Q8EAD9oEnK2lMgmhIKC2E= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1o_9BqNra6zh_HmLvWFNQm4WYzN2W1bygUabZEe9Lzsg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 78080 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 78080 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1KePfYaceZ29tiGigvOtcCFJNsP7pueMsjzyHKVb7w64= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16tOU8-XLYrNPdvvuRaPZblzPTCq0IYpmOqRVYwuw2zIA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 27904 + LdsInitCVgprs: false + LdsNumBytes: 27904 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27904 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16iQaL89dfDqFaUwHSFuqBVL1d7gxxRqiMEISWe48rBhk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16G697WmrCkkuSW0ygicOEsjSabf2uABvGDaKBNngK4oE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 12544 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 12544 + LdsOffsetB_Blk: 45312 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 45312 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1opMCZdZFzR31a1nXMExXx-HFGwt6FHBNysNBuXPo34o= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1GPhaxX2PStJ8QfX3S-AgQFNJD6o4d_c2qqsOdt_oaC4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 44288 + LdsInitCVgprs: false + LdsNumBytes: 44288 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44288 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 7 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1gqbADNhoP3bHba8oMSfqNYB-76ixvzWU8E58C_oXxBo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1-VJoMMTo08ubflQERsCoO0j_TzMOoFOBMG1VV9dqzpY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI15OARqHSqH8kbGBPru-B8g6Oad-Zo3FGsUS8SxkzZZYk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16xVRyCL8pcYWjhmBX3Sksf7_ECJZKzPUO0YM7WCFd9sM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 23808 + LdsInitCVgprs: false + LdsNumBytes: 23808 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23808 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI162emMCIrLY9uAeJtwZl6HlANVDofIksu3xpgoJdG4L7I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16UNManzuffvMTmsgfWnWF6Bt5jjvtS0tBsdToDQ_bZvc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 128 + LSPA: 64 + LSPB: 32 + LVCA: 4 + LVCB: 8 + LVPA: 4 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 24832 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 8448 + LdsOffsetMetadata_Blk: 24832 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1kHn9o1wGnb0Le03rKuL4bApCYLkFlSLtoU6MjkRibzc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1NMWe8o4PG1BIiKP1ggEjZUpwkkJSDyvBBNMZ8tXiaR8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 69888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 69888 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 1 + NumLoadsB: 7 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1Q0rm-ZJ-B1Ep3tDRV7Y6v5xFppIw1pc1wB6W7qCB6eE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 32000 + LdsInitCVgprs: false + LdsNumBytes: 32000 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32000 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI14_XY01zUqMSgVV3Z29zc0HvqY2RBU1XaMixpmKPTFvQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1BV068w-wQamiyACMJrCQO6AzvkFtVVaL6YUOIybq_As= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI167ogjNu2arPxWNgqgKM8sDxj_QbeHe5G3biyDU4kITPE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 19712 + LdsInitCVgprs: false + LdsNumBytes: 19712 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 37120 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 19712 + LdsOffsetMetadata_Blk: 37120 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI1606uPSrg8jSF_adB8-Q1EUFxmHwf5F9lLMz64s3UmyCk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16Jm6coo6HlXoYc23v3SkS10L-uM9lwa5bnHnY2Vs_oLo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 8 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 20736 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 4352 + LdsOffsetMetadata_Blk: 20736 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIv9D6qCD557UuGj0WYoB1KLme1Sh1mzoR8zvP-Ew1rDY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 66560 + LdsInitCVgprs: false + LdsNumBytes: 66560 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 66560 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 13] + MIWaveTileA: 4 + MIWaveTileB: 13 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 208 + MacroTileA: 256 + MacroTileB: 208 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 52 + NumLoadsA: 8 + NumLoadsB: 13 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 13 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 13 + ThreadTileA: 16 + ThreadTileB: 13 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIZTzgVMYbF8zAjkLbWPx_KguNjRPGDWzQphgfwu106mI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 28160 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 11] + MIWaveTileA: 4 + MIWaveTileB: 11 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 176 + MacroTileA: 256 + MacroTileB: 176 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 44 + NumLoadsA: 8 + NumLoadsB: 11 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 11 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 11 + ThreadTileA: 16 + ThreadTileB: 11 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIYpXgEPF2KmPhp_HSItkaLdu8ZfTql6wuzRMctpM8bWs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 121856 + LdsInitCVgprs: false + LdsNumBytes: 121856 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33280 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 9] + MIWaveTileA: 4 + MIWaveTileB: 9 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 144 + MacroTileA: 256 + MacroTileB: 144 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 8 + NumLoadsB: 9 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 9 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 9 + ThreadTileA: 16 + ThreadTileB: 9 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIM6ky9Tq6VRezat1v-Vy4Qy4OEEUGyFJHwE-XvlIpRaY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 17920 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 112 + MacroTileA: 256 + MacroTileB: 112 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1ZaT28UTzCJBM6wCUWArJk9cu4afCdA6oGhULOQC7jwQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 46080 + LdsInitCVgprs: false + LdsNumBytes: 46080 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 12800 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46080 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1ckLt2U3Q4BSlwSpNR-cpd6uJ3sMr5OVPRWuKvEiSVtY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 7680 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB128_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1Cks6yoWu7mRHY5OU_L0wlyt_BkdBZ9B6KYLX0jMjEZM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 35840 + LdsInitCVgprs: false + LdsNumBytes: 35840 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 2560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1280_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA5_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1LRieWm8obTv5BrugK6ZfaH9J4-zT3LrRBcx6GxwPdUk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 1280 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45312 + LdsInitCVgprs: false + LdsNumBytes: 45312 + LdsNumElementsAlignedA: 10496 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 10496 + LdsOffsetB_Blk: 76032 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45312 + LdsOffsetMetadata_Blk: 76032 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsCoalescedA: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA768_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1zBVp9O_j-pDQmFJ0pruVZjPjyXjbaxIDNKuOKzE-hUk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 768 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 6400 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 6400 + LdsOffsetB_Blk: 71936 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 71936 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1LmEbjkFjRe2EHvxZ5njVJSBpnw78dqd37eC05zoq8oY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 128 + LSPA: 128 + LSPB: 32 + LVCA: 2 + LVCB: 8 + LVPA: 16 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 2304 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2304 + LdsOffsetB_Blk: 67840 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 67840 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB1024_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI_Cpsqq7IdCcI6m09ixnRrSz_HTRJbwY41LkpuZ6eaow= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1tP0yLw5sOaqSqLw69T0rUISjIec4gtRcEBnLindXxd0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB512_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1WlKSkQYWc45q9-9u3Sa8KR5G5YSwdMXXvYIooja5mzU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1ZFz2Wqi9AGPfO9YZeT8hu0U-a6QRpBHZ1MNZhOuWfzs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1536_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA3_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16YTIDOkvWpHTvJMbXMK3Qn2TUcHpLDIz3L67zYqwLtxM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 1536 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 24832 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 24832 + LdsOffsetB_Blk: 90368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 90368 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsCoalescedA: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1Hyiuad9E_LwogHYOQgHONBkR83eXTUE_-HwHw1z6c2M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 62976 + LdsInitCVgprs: false + LdsNumBytes: 62976 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 62976 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 10 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1_vkhGjNt0PNt_xV8VVOz2XzUCfwEcU8zljVlbKTICVg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16riuLixhIaRRQRAFH27-FM-5iTnYrSUKGh2HFdVNH4W8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16tRLsgniRf9NQ8cwUknJpPeoosk2I0euxrG_yXuRD6Zc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16zCjLhPMzebcvjlA5OUrujTBGN1OYulB3ktj1a570vLs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 256 + LSPA: 64 + LSPB: 16 + LVCA: 4 + LVCB: 16 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1OuevmgTEfRQM80_r7hzorlLPNIZiSHCEk0LhjgcOAH8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 60672 + LdsInitCVgprs: false + LdsNumBytes: 60672 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 60672 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 12 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI117HfbCeMXp_75IhCNgyVsAtudVa47Y080duX4tlb_Cc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 54528 + LdsInitCVgprs: false + LdsNumBytes: 54528 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 54528 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 10 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1w_TVfNrjE3mYvq96Bm1Jgkl-HCbXvzNQ6H6gr4PSnKc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16IX89BCLJtYzlNb-bQG51jUiesgW7gn9abBGKzaC4u7Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 73984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 73984 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16yEC0PHvmD0zrKvRDTkhqsmiXq78VvD0Eg5fcAnocoic= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16_OtB8fTrZk3Pt0bE3bXfsWfkoWXEPa2ENPfe4eaiVRI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 256 + LSPA: 128 + LSPB: 16 + LVCA: 2 + LVCB: 16 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 8448 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8448 + LdsOffsetB_Blk: 41216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41216 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1uaMQi1863lUOUemwTE99mttXxpTRkvqvmiwvtfx6Qx0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 80384 + LdsInitCVgprs: false + LdsNumBytes: 80384 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 13824 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 80384 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 16 + NumLoadsB: 3 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA4096_LBSPPB256_LBSPPM0_LPA64_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1_aftRJpk6u1tTs3-Hwls-Fo3A3dD889Os-hWuvfNQvw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 4096 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 71168 + LdsInitCVgprs: false + LdsNumBytes: 71168 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 4608 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71168 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 64 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI18CjcSziK0PjBic2MAopWR-Ta_nHbDse6qNtds5pye7Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 256 + LSPA: 256 + LSPB: 16 + LVCA: 1 + LVCB: 16 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 71936 + LdsInitCVgprs: false + LdsNumBytes: 71936 + LdsNumElementsAlignedA: 4352 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4352 + LdsOffsetB_Blk: 135424 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71936 + LdsOffsetMetadata_Blk: 135424 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 512 + LSPA: 64 + LSPB: 8 + LVCA: 4 + LVCB: 32 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI169OaDNKWZnXMa1R_AlPEgQpVqAFVb54TBTn3PvUMxY7A= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 512 + LSPA: 128 + LSPB: 8 + LVCA: 2 + LVCB: 32 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16JRL6UviIw3ql2MWJlkjDhAPVUQuM-dfo_YP9V8INeCE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16a9rQBxUkyvDWlmJquyIGWqzMINaNpRD_aWL9QWIuzQ8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 512 + LSPA: 128 + LSPB: 8 + LVCA: 2 + LVCB: 32 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 49664 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 49664 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16Jyq1ax9kWbpfAk08jNUVf7z1XoXRVCVdpt-kBpa0fWY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 64 + LSCB: 512 + LSPA: 64 + LSPB: 8 + LVCA: 4 + LVCB: 32 + LVPA: 4 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 98816 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 98816 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16D0scrs9_FOYkVHUHqfaWIBeu8Rz72d3pOV3R8iV8H74= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI169Ao_eCjaajChoPx9i0y3MGqMyeJdbqU3dbjwB_UIKx4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 512 + LSPA: 256 + LSPB: 8 + LVCA: 1 + LVCB: 32 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1DDaFTcewwV5Ybn8swfgQtfH13aPb98dLOnRrTIJYmSQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1ifWE51xli667eSWHdf8X0Gqm1isofKMCZj08QiVl3U0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 1024 + LSPA: 128 + LSPB: 4 + LVCA: 2 + LVCB: 64 + LVPA: 8 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1U06udU6pkFuvB0exyiBHs-wOgMdXOgbcv4nENk4uw44= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA16_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1oDw0RAqdNEFhP7idqV0O87totk9ugWGoiXKFb7JRQVw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: false + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 16 + LSCB: 1024 + LSPA: 256 + LSPB: 4 + LVCA: 1 + LVCB: 64 + LVPA: 16 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84992 + LdsInitCVgprs: false + LdsNumBytes: 84992 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84992 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 16 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 16 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA16_LPB8_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Ailk_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xqwn4mYe9PqjtRIapRznDsflXKMnaPO3v98RcshyCRoc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 16 + LdsPadB: 8 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: false + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: false + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index 48b591a0c796..7a88c80a39d4 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,19809 +81,14189 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI93W5TiEccmFg8wGd3V3_i7anSx2in1g2Nvgr1nxFMW0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1WBDB2zkUie8-CEEoI_ArcQdbGzpndP9LCMEirfw4oOM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46336 - LdsInitCVgprs: false - LdsNumBytes: 46336 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46336 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1byiwSEFKpQOHfD3AL3aMNi54Gti0tPrWia0W_RI3F9I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1drE7ijfboWVMJx6XbcD7R5giTiA773IMYwA0SWdbSew= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIlwPgpsfEr1tHJhWc_dQ16xuUglNHBRvb9FY8kgrV2uo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 128 - MacroTileA: 224 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1vpzVkbJ5WB1luwHLUPLp6GT1RJDqsKYFBruB0CuPp3k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48384 - LdsInitCVgprs: false - LdsNumBytes: 48384 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48384 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 3] - MIWaveTileA: 7 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 96 - MacroTileA: 224 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 7 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 3 - ThreadTileA: 28 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1zuV98IyOwxElPoy5ichUD4fa6nssRrdqd1Q3modyHdk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44288 - LdsInitCVgprs: false - LdsNumBytes: 44288 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44288 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 7 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1xqlNtEYE-E0Mz0QATa9_dzumE9-a3Wvr1cR9tzlzYW8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 7 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICgefWmv3HWofIYKXJsq8855MHgkupzgrkVMcs4u-by0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 125952 - LdsInitCVgprs: false - LdsNumBytes: 125952 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 8] - MIWaveTileA: 6 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 256 - MacroTileA: 192 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 8 - ThreadTileA: 24 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MInHiNZ23NKfFsK7oWehXsdVhdC3keTzJy0tZ_ibW9pzc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1rAFgL4PTZQ2gx_CQR3J0YNYjnR6ZpZqLT3kKoVcsb8Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1Vg4WjQQgNmrK6Ytw7ZwOj3TBnXGNte5TZoyB4kGGvxQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1WRxklg9QBBuMW_MrlGv_hyeD_1KjPxvE5CvBb2xGGdk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32000 - LdsInitCVgprs: false - LdsNumBytes: 32000 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 60416 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32000 - LdsOffsetMetadata_Blk: 60416 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MIGmZq9BXDiHJ0tc5zxXOziePTEBPwz6cUZrAUknIRT7E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123904 - LdsInitCVgprs: false - LdsNumBytes: 123904 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 8] - MIWaveTileA: 5 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 256 - MacroTileA: 160 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 160 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 8 - ThreadTileA: 20 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIwjK4OAX-B9dLgo7aLraklJg0ipZLFjGXad70tFlHa_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50432 - LdsInitCVgprs: false - LdsNumBytes: 50432 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50432 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 6] - MIWaveTileA: 5 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 192 - MacroTileA: 160 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 120 - NumLoadsA: 5 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 6 - ThreadTileA: 20 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIXGlL2yJ1nYGKMcvOKU86eOyg9fj72cyU6dwaNXql1aE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1d_LfxY9vu-XLyyQyjblMOjGGccv6SWDTcAMP2nTGrZ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1xJMqskwq5v3yaRA45HS_l-Q0PeuwEvXnjkyc8iEsIuQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1J2YusILFQCwZJTPycybXWTPnKyMth_wvsZJGJUL6XBA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 58368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 58368 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIK3HX_ARTjdrJyQgn8ka1A08cR0NqiFPRF8k7JYx3lno= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50176 - LdsInitCVgprs: false - LdsNumBytes: 50176 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50176 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 8] - MIWaveTileA: 4 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 8 - ThreadTileA: 16 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIAYYyDwfBJqmmCRgY0wORDPvjgSZIA73oaclExRH_3Ks= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIL86z76qmkuTn8D76ItocU1Fqkm1UTu79IZpG8k_14Mo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38144 - LdsInitCVgprs: false - LdsNumBytes: 38144 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38144 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MIcEWO-iBcRFoqXFm5FnHIJnGhh114ULGYJcbTf6WzdFI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1uj-yswGYT2k5VXIiR3VHictFIePTXmx0STeF7oDnJ6c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1_uypCl737WTznRBFG8qQSTk3KV2VA62EV3JnGAPhybs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1SN9OiW0wmLBhcMdgMSszjYfbUYfIwyxoyCZVswMyYtQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1OdZvyKg_z2Mm3twNkO33DsDFI4XHPhixfFGT8M7-fns= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 48128 - LdsInitCVgprs: false - LdsNumBytes: 48128 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 48128 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1-rYnfwe4wvrEHrwuVO2V79PzVjbfHc2PuGqRKVJDgfI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40192 - LdsInitCVgprs: false - LdsNumBytes: 40192 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40192 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1QbCX-NbN9_FulNkEv2WcOQVCSkdBBRKzE6PDyeD0PIc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI15g6oi8kYRfTdSkq4hb7l4vHYA3S6a_b3gQ32PTizD-c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32256 - LdsInitCVgprs: false - LdsNumBytes: 32256 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32256 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16zaeoUJs0j0gHGLtiiAi_mKOg4oYgz97kp3_f949MqD0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 27904 - LdsInitCVgprs: false - LdsNumBytes: 27904 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27904 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16n1I05diAnU09ahBMg4ZOJ69DUU98umJt7phil649P_I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 23808 - LdsInitCVgprs: false - LdsNumBytes: 23808 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23808 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16qYVsRYavL1_uEB81mfehkE0xypBMpwpmv8NTKwWUCZQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 19712 - LdsInitCVgprs: false - LdsNumBytes: 19712 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 19712 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1rLkRPc5Fv6KcZ0pkpjti_dDNkwatHGJAjTt5YgSUF0A= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1MsUtatzu8gT7lwtRqOR1GlkYA3LAi1h7-WpiR8njS-s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI17ixososrZhbGip5glmYhzY0dhoWQ6s92NzkMuYpNd6c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1O6oTiyV66Nmoj0i8-AQlJj5R-NLtNsfH6diaQrWbGzQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16BDy2huNg1pYAEVpwRmCSx6TZ3y-fyTBhxwPrXBTUMn0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 21760 - LdsInitCVgprs: false - LdsNumBytes: 21760 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 21760 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16T76NxLk5ebc1y9bf0bGaRoLE1Btglvx1amB35nt2jrg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI164mp4kBtFpiXW7TfMmUnYxQd7fjnanNrZue8IY8R1Nt4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29824 - LdsInitCVgprs: false - LdsNumBytes: 29824 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 25600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 9216 - LdsOffsetMetadata_Blk: 25600 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1CBdovKQljtt9JJeN35QUWWaGXoZh_9PeLDfb3HuWMiA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37888 - LdsInitCVgprs: false - LdsNumBytes: 37888 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 32768 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37888 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 0 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1eYnh8ExIG9i1WyAXPWhGKC2q4srWo55-2vcORDH9kEs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 3072 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 29952 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1m1tiKe28X_8BSOyU__5gEgsr8TQY2h1rX5gF4hmE4Qc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2560 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 20736 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI17VRJH_5pnmfoUU-FHr32iGMlOdeQUYJplapfD41zqrY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22016 - LdsInitCVgprs: false - LdsNumBytes: 22016 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22016 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16AqtF9WEJ-EpQbhJnOZKbC372wZ4o2kh1rPtJZau1G1U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 12544 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16WqqFFjCIevU2bTOwvq53GHxOVUs4xuB_iw0OyIMRO9s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 64 - LSPA: 32 - LSPB: 64 - LVCA: 8 - LVCB: 4 - LVPA: 2 - LVPB: 4 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 29952 - LdsInitCVgprs: false - LdsNumBytes: 29952 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16O0pSqsQ7I3lGvs23ES_yKR_wzzq3HSwM1IcyJmGtnhc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 32 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25728 - LdsInitCVgprs: false - LdsNumBytes: 25728 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1Jybduy0hX_yQp-cMS6QlHSWe9YueMgbIOjgYTfyE4xY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1280 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45312 - LdsInitCVgprs: false - LdsNumBytes: 45312 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 10496 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45312 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 5 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1Mhf-Vi17l2Fosir59CVL3Da6ElGCICe4ioDdRimtZBI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 768 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41216 - LdsInitCVgprs: false - LdsNumBytes: 41216 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 6400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41216 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1zfbotELeA8YVwjd3HOC5VjWft_ZNaVArBeo8Zgzp6wo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 16 - LSPA: 32 - LSPB: 128 - LVCA: 8 - LVCB: 2 - LVPA: 2 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37120 - LdsInitCVgprs: false - LdsNumBytes: 37120 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 2304 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37120 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIPJDSEwAxzWM4fDuAEI3dit5Qm8ipFSRMSAIZnl6eve4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 66560 - LdsInitCVgprs: false - LdsNumBytes: 66560 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 66560 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [13, 4] - MIWaveTileA: 13 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 208 - MacroTile1: 256 - MacroTileA: 208 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 208 - NumLoadsA: 13 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 13 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 52 - ThreadTile1: 4 - ThreadTileA: 52 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIHlGj0Igh78VESckoEVhEEojHcEyAVYdCerN_cX6Ldrw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 28160 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 28160 - LdsOffsetB_Blk: 93696 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 28160 - LdsOffsetMetadata_Blk: 93696 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [11, 4] - MIWaveTileA: 11 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 176 - MacroTile1: 256 - MacroTileA: 176 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 176 - NumLoadsA: 11 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 11 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 44 - ThreadTile1: 4 - ThreadTileA: 44 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MIs43vjb8vXLGfeBgy4ohTkGSxXW8PsCp3uz_zeqpVxeM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 121856 - LdsInitCVgprs: false - LdsNumBytes: 121856 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 88576 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23040 - LdsOffsetMetadata_Blk: 88576 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [9, 4] - MIWaveTileA: 9 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 144 - MacroTile1: 256 - MacroTileA: 144 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 144 - NumLoadsA: 9 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 9 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 36 - ThreadTile1: 4 - ThreadTileA: 36 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIYY998OJmuRcBnYy6KGEItwDq2j2QehUOs8VfioGt_EY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17920 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17920 - LdsOffsetB_Blk: 83456 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 83456 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 112 - MacroTile1: 256 - MacroTileA: 112 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1riWq2NrSVGbHY4ulT3CBZ1ka9jZSPImqGDsOLfR739U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 46080 - LdsInitCVgprs: false - LdsNumBytes: 46080 - LdsNumElementsAlignedA: 12800 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12800 - LdsOffsetB_Blk: 78336 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 46080 - LdsOffsetMetadata_Blk: 78336 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI109tIWmrNFpIMCvCKD44FEjLK6mNrPP0ecMYRKJWv__o= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 7680 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 7680 - LdsOffsetB_Blk: 73216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 73216 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1ePlkOMYOjA6Sj3U5aK9R-vdkoYkQzZKOf49wW2bLE70= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 2 - LVPB: 1 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 35840 - LdsInitCVgprs: false - LdsNumBytes: 35840 - LdsNumElementsAlignedA: 2560 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2560 - LdsOffsetB_Blk: 68096 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 68096 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1R3pW5TGY8A5opfshY0CC3g5uD4DXiGRwS1VnQHSdtTs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 60672 - LdsInitCVgprs: false - LdsNumBytes: 60672 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 117760 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 60672 - LdsOffsetMetadata_Blk: 117760 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI13Sbf9PQEx5F-iKE9julI7_PdDt3yPapFP4vCFUpDo1E= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 62976 - LdsInitCVgprs: false - LdsNumBytes: 62976 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 62976 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 10 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI137t5Uvc0ssl50sE4SNlCcUp-XUoCcLX9_I4uGMXledk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 54528 - LdsInitCVgprs: false - LdsNumBytes: 54528 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 54528 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 10 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIOmRHwGdcC8gFOr3dPzasFfQWs0PP2cDcwy6aKZU2f5U= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1454Bwxd_MIvQjuG-Y_CvHs_qnR7PgRRvHGhCUO-XbOY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1Rmqg4myY3vDAMf_l0hT_13f8mLa__KusFSpIvX00Gq4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42240 - LdsInitCVgprs: false - LdsNumBytes: 42240 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42240 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI15NEBMkxG3p9Ig0Cr9U7Vscm7bIb_z6qr2ylZCFd2y30= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16wxJkzA64nljgT6A4tMYPR-qnvL3NK7-LGLeTrL93K28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 44544 - LdsInitCVgprs: false - LdsNumBytes: 44544 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 44544 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16s4XHH5Mm_pX3J5f5thK2zPFz4wk3wfVK1ZsVsKy_LpE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36096 - LdsInitCVgprs: false - LdsNumBytes: 36096 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36096 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1XeU_kq-TCV7gFRA0zjadiPg5xVzK4J_9tcfhp6RHcE8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16uy_AuCnsXVT_W8ZUwPY2xZhbdvr0s1-k6m6FDS2GyPc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16JHWtGGP8udHdAtFC_ITt55YeNZjw7qy9YbaDQmBmLnw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25856 - LdsInitCVgprs: false - LdsNumBytes: 25856 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25856 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1CMycMeJchCKewZBbLPkfdDttQ7jpxPNi2DcgEi0aA7Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 1 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI167PfjxLOUHYcmsWqM3Jr5QHvep_HsB981kR15bVmyR_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1536 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34048 - LdsInitCVgprs: false - LdsNumBytes: 34048 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 24832 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34048 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 3 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16hW4G8jnTFUiWZWnW0Tmfmi9zpySUXkWHngzRijqsswA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 64 - LSPA: 16 - LSPB: 64 - LVCA: 16 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16pkwZPXEjbcu5SfItYZgOkhxquryDik7C3Mlw6DhdC9c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 32 - LSPA: 16 - LSPB: 128 - LVCA: 16 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17664 - LdsInitCVgprs: false - LdsNumBytes: 17664 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 8448 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17664 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1ag0qOUjDoYcyoioR4OdNEh8L5HMCVTkf-1kVLG0Ke8Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 16 - LSPA: 16 - LSPB: 256 - LVCA: 16 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71936 - LdsInitCVgprs: false - LdsNumBytes: 71936 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 4352 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71936 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1RElBQ1lA5tfxLjjMGZY4_t9WElkqvIrWw-ZvypHxsS4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 80384 - LdsInitCVgprs: false - LdsNumBytes: 80384 - LdsNumElementsAlignedA: 13824 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 13824 - LdsOffsetB_Blk: 144896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 80384 - LdsOffsetMetadata_Blk: 144896 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1g4O7XQYq5Erjhq0mbnd0ZzsOn7Y74NZ0Ed50NCU8Cko= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 4096 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71168 - LdsInitCVgprs: false - LdsNumBytes: 71168 - LdsNumElementsAlignedA: 4608 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4608 - LdsOffsetB_Blk: 135680 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71168 - LdsOffsetMetadata_Blk: 135680 - LdsPadA: 32 - LdsPadB: 64 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 64 - LSPA: 8 - LSPB: 64 - LVCA: 32 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16c-3ymAqE5iIzAVIl4M4mFas_yKBzwB8S5hIJHrx_G10= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 32 - LSPA: 8 - LSPB: 128 - LVCA: 32 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16l2koiE9edxgF5EcPYxeXQhdktTUf3O_Nh5I8-rOv_5c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16PDTjUmNdBPeXIoutsSh54JIRkDgOgTuGJ5_UUnXYC30= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16FZJXxZzBUYpkHY1uHTOo5xfvmHSftJ7Ss8JFmsnHOq4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 16 - LSPA: 8 - LSPB: 256 - LVCA: 32 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16UExxzleg69jKCOIPv3aHKRuxpeCJ0OtN8U8_i8380wE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 32 - LSPA: 8 - LSPB: 128 - LVCA: 32 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 25600 - LdsInitCVgprs: false - LdsNumBytes: 25600 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16486QKgKm2Q-ueCL9nw3616pMoSPYJwy96wpaKy2ZR9g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 64 - LSPA: 8 - LSPB: 64 - LVCA: 32 - LVCB: 4 - LVPA: 1 - LVPB: 4 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 41984 - LdsInitCVgprs: false - LdsNumBytes: 41984 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 41984 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1gJ-iaCReTZwgtOaq-r_7Ae1Xx5lpHHILXpRm4hXIFWo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34304 - LdsInitCVgprs: false - LdsNumBytes: 34304 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34304 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1TOymj1f06yNxQYQNnDElBwV9Je3kDkNSQdL4ZFlGHxw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1eVyecsmYTDI_tnvIi0T4aak3HoYfyCs1_BMjsWhSVGA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 16 - LSPA: 4 - LSPB: 256 - LVCA: 64 - LVCB: 1 - LVPA: 1 - LVPB: 16 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84992 - LdsInitCVgprs: false - LdsNumBytes: 84992 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84992 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1SvhFCnFpOT5VXBaOtUT022uRF_6hTz1blD7UqHf2kpA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: false - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 32 - LSPA: 4 - LSPB: 128 - LVCA: 64 - LVCB: 2 - LVPA: 1 - LVPB: 8 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xDWExf5YqS6e0I09q-b7LtSPBEclPiz1pvMgNxDpD5yY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 32 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 1 - GlobalReadVectorWidthB: 1 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 32 - LSCB: 32 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 8 - LVPB: 8 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 6464 - LdsInitCVgprs: false - LdsNumBytes: 6464 - LdsNumElementsAlignedA: 1280 - LdsNumElementsAlignedB: 1280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 4096 - LdsOffsetB: 1280 - LdsOffsetB_Blk: 5376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 1280 - LdsOffsetMetadata_Blk: 5376 - LdsPadA: 8 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 8 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 32 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 32, 1, 1, 1] - MIInputPerThread: 8 - MIInputPerThreadA: 8 - MIInputPerThreadB: 8 - MIInputPerThreadMetadata: 8 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 32 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 32, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchOpt: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: false - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 32 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: false - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: -1 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 32 - _DepthUA: 32 - _DepthUB: 32 - _DepthUMetadata: 32 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 1 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 3 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI93W5TiEccmFg8wGd3V3_i7anSx2in1g2Nvgr1nxFMW0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1WBDB2zkUie8-CEEoI_ArcQdbGzpndP9LCMEirfw4oOM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 46336 + LdsInitCVgprs: false + LdsNumBytes: 46336 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46336 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI1byiwSEFKpQOHfD3AL3aMNi54Gti0tPrWia0W_RI3F9I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI1drE7ijfboWVMJx6XbcD7R5giTiA773IMYwA0SWdbSew= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIlwPgpsfEr1tHJhWc_dQ16xuUglNHBRvb9FY8kgrV2uo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 128 + MacroTileA: 224 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 4 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1vpzVkbJ5WB1luwHLUPLp6GT1RJDqsKYFBruB0CuPp3k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 48384 + LdsInitCVgprs: false + LdsNumBytes: 48384 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48384 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 3] + MIWaveTileA: 7 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 96 + MacroTileA: 224 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 7 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 3 + ThreadTileA: 28 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1zuV98IyOwxElPoy5ichUD4fa6nssRrdqd1Q3modyHdk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 44288 + LdsInitCVgprs: false + LdsNumBytes: 44288 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44288 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 7 + NumLoadsB: 2 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1xqlNtEYE-E0Mz0QATa9_dzumE9-a3Wvr1cR9tzlzYW8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 7 + NumLoadsB: 1 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICgefWmv3HWofIYKXJsq8855MHgkupzgrkVMcs4u-by0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 125952 + LdsInitCVgprs: false + LdsNumBytes: 125952 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 8] + MIWaveTileA: 6 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 256 + MacroTileA: 192 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 8 + ThreadTileA: 24 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MInHiNZ23NKfFsK7oWehXsdVhdC3keTzJy0tZ_ibW9pzc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1rAFgL4PTZQ2gx_CQR3J0YNYjnR6ZpZqLT3kKoVcsb8Y= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1Vg4WjQQgNmrK6Ytw7ZwOj3TBnXGNte5TZoyB4kGGvxQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1WRxklg9QBBuMW_MrlGv_hyeD_1KjPxvE5CvBb2xGGdk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 32000 + LdsInitCVgprs: false + LdsNumBytes: 32000 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 60416 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32000 + LdsOffsetMetadata_Blk: 60416 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MIGmZq9BXDiHJ0tc5zxXOziePTEBPwz6cUZrAUknIRT7E= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 123904 + LdsInitCVgprs: false + LdsNumBytes: 123904 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 8] + MIWaveTileA: 5 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 256 + MacroTileA: 160 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 160 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 8 + ThreadTileA: 20 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIwjK4OAX-B9dLgo7aLraklJg0ipZLFjGXad70tFlHa_4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 50432 + LdsInitCVgprs: false + LdsNumBytes: 50432 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50432 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 6] + MIWaveTileA: 5 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 192 + MacroTileA: 160 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 120 + NumLoadsA: 5 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 6 + ThreadTileA: 20 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIXGlL2yJ1nYGKMcvOKU86eOyg9fj72cyU6dwaNXql1aE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1d_LfxY9vu-XLyyQyjblMOjGGccv6SWDTcAMP2nTGrZ4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1xJMqskwq5v3yaRA45HS_l-Q0PeuwEvXnjkyc8iEsIuQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1J2YusILFQCwZJTPycybXWTPnKyMth_wvsZJGJUL6XBA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 58368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 58368 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MIK3HX_ARTjdrJyQgn8ka1A08cR0NqiFPRF8k7JYx3lno= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 50176 + LdsInitCVgprs: false + LdsNumBytes: 50176 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50176 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 8] + MIWaveTileA: 4 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 8 + ThreadTileA: 16 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIAYYyDwfBJqmmCRgY0wORDPvjgSZIA73oaclExRH_3Ks= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MIL86z76qmkuTn8D76ItocU1Fqkm1UTu79IZpG8k_14Mo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 38144 + LdsInitCVgprs: false + LdsNumBytes: 38144 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38144 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MIcEWO-iBcRFoqXFm5FnHIJnGhh114ULGYJcbTf6WzdFI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1uj-yswGYT2k5VXIiR3VHictFIePTXmx0STeF7oDnJ6c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1_uypCl737WTznRBFG8qQSTk3KV2VA62EV3JnGAPhybs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1SN9OiW0wmLBhcMdgMSszjYfbUYfIwyxoyCZVswMyYtQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1OdZvyKg_z2Mm3twNkO33DsDFI4XHPhixfFGT8M7-fns= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 48128 + LdsInitCVgprs: false + LdsNumBytes: 48128 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 48128 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1-rYnfwe4wvrEHrwuVO2V79PzVjbfHc2PuGqRKVJDgfI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 40192 + LdsInitCVgprs: false + LdsNumBytes: 40192 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40192 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1QbCX-NbN9_FulNkEv2WcOQVCSkdBBRKzE6PDyeD0PIc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI15g6oi8kYRfTdSkq4hb7l4vHYA3S6a_b3gQ32PTizD-c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 32256 + LdsInitCVgprs: false + LdsNumBytes: 32256 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32256 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16zaeoUJs0j0gHGLtiiAi_mKOg4oYgz97kp3_f949MqD0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 27904 + LdsInitCVgprs: false + LdsNumBytes: 27904 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27904 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16n1I05diAnU09ahBMg4ZOJ69DUU98umJt7phil649P_I= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 23808 + LdsInitCVgprs: false + LdsNumBytes: 23808 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23808 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16qYVsRYavL1_uEB81mfehkE0xypBMpwpmv8NTKwWUCZQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 19712 + LdsInitCVgprs: false + LdsNumBytes: 19712 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 19712 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI1rLkRPc5Fv6KcZ0pkpjti_dDNkwatHGJAjTt5YgSUF0A= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1MsUtatzu8gT7lwtRqOR1GlkYA3LAi1h7-WpiR8njS-s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI17ixososrZhbGip5glmYhzY0dhoWQ6s92NzkMuYpNd6c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1O6oTiyV66Nmoj0i8-AQlJj5R-NLtNsfH6diaQrWbGzQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16BDy2huNg1pYAEVpwRmCSx6TZ3y-fyTBhxwPrXBTUMn0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 21760 + LdsInitCVgprs: false + LdsNumBytes: 21760 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 21760 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16T76NxLk5ebc1y9bf0bGaRoLE1Btglvx1amB35nt2jrg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI164mp4kBtFpiXW7TfMmUnYxQd7fjnanNrZue8IY8R1Nt4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 29824 + LdsInitCVgprs: false + LdsNumBytes: 29824 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 25600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 9216 + LdsOffsetMetadata_Blk: 25600 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB0_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1CBdovKQljtt9JJeN35QUWWaGXoZh_9PeLDfb3HuWMiA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 37888 + LdsInitCVgprs: false + LdsNumBytes: 37888 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 32768 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37888 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 0 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB3072_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1eYnh8ExIG9i1WyAXPWhGKC2q4srWo55-2vcORDH9kEs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 3072 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 29952 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2560_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI1m1tiKe28X_8BSOyU__5gEgsr8TQY2h1rX5gF4hmE4Qc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2560 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 20736 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI17VRJH_5pnmfoUU-FHr32iGMlOdeQUYJplapfD41zqrY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 22016 + LdsInitCVgprs: false + LdsNumBytes: 22016 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22016 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16AqtF9WEJ-EpQbhJnOZKbC372wZ4o2kh1rPtJZau1G1U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 12544 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16WqqFFjCIevU2bTOwvq53GHxOVUs4xuB_iw0OyIMRO9s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 64 + LSPA: 32 + LSPB: 64 + LVCA: 8 + LVCB: 4 + LVPA: 2 + LVPB: 4 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 29952 + LdsInitCVgprs: false + LdsNumBytes: 29952 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16O0pSqsQ7I3lGvs23ES_yKR_wzzq3HSwM1IcyJmGtnhc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 32 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25728 + LdsInitCVgprs: false + LdsNumBytes: 25728 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1280_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB5_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1Jybduy0hX_yQp-cMS6QlHSWe9YueMgbIOjgYTfyE4xY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1280 + LdsBytesNoAmax: 45312 + LdsInitCVgprs: false + LdsNumBytes: 45312 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 10496 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45312 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsCoalescedB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB768_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1Mhf-Vi17l2Fosir59CVL3Da6ElGCICe4ioDdRimtZBI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 768 + LdsBytesNoAmax: 41216 + LdsInitCVgprs: false + LdsNumBytes: 41216 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 6400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41216 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1zfbotELeA8YVwjd3HOC5VjWft_ZNaVArBeo8Zgzp6wo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 16 + LSPA: 32 + LSPB: 128 + LVCA: 8 + LVCB: 2 + LVPA: 2 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 37120 + LdsInitCVgprs: false + LdsNumBytes: 37120 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 2304 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37120 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIPJDSEwAxzWM4fDuAEI3dit5Qm8ipFSRMSAIZnl6eve4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 66560 + LdsInitCVgprs: false + LdsNumBytes: 66560 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 66560 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [13, 4] + MIWaveTileA: 13 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 208 + MacroTile1: 256 + MacroTileA: 208 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 208 + NumLoadsA: 13 + NumLoadsB: 8 + NumLoadsPerpendicularA: 13 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 52 + ThreadTile1: 4 + ThreadTileA: 52 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIHlGj0Igh78VESckoEVhEEojHcEyAVYdCerN_cX6Ldrw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 28160 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 28160 + LdsOffsetB_Blk: 93696 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 28160 + LdsOffsetMetadata_Blk: 93696 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [11, 4] + MIWaveTileA: 11 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 176 + MacroTile1: 256 + MacroTileA: 176 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 176 + NumLoadsA: 11 + NumLoadsB: 8 + NumLoadsPerpendicularA: 11 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 44 + ThreadTile1: 4 + ThreadTileA: 44 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MIs43vjb8vXLGfeBgy4ohTkGSxXW8PsCp3uz_zeqpVxeM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 121856 + LdsInitCVgprs: false + LdsNumBytes: 121856 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 88576 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23040 + LdsOffsetMetadata_Blk: 88576 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [9, 4] + MIWaveTileA: 9 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 144 + MacroTile1: 256 + MacroTileA: 144 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 144 + NumLoadsA: 9 + NumLoadsB: 8 + NumLoadsPerpendicularA: 9 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 36 + ThreadTile1: 4 + ThreadTileA: 36 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIYY998OJmuRcBnYy6KGEItwDq2j2QehUOs8VfioGt_EY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17920 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17920 + LdsOffsetB_Blk: 83456 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 83456 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 112 + MacroTile1: 256 + MacroTileA: 112 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1riWq2NrSVGbHY4ulT3CBZ1ka9jZSPImqGDsOLfR739U= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 46080 + LdsInitCVgprs: false + LdsNumBytes: 46080 + LdsNumElementsAlignedA: 12800 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12800 + LdsOffsetB_Blk: 78336 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 46080 + LdsOffsetMetadata_Blk: 78336 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI109tIWmrNFpIMCvCKD44FEjLK6mNrPP0ecMYRKJWv__o= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 7680 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 7680 + LdsOffsetB_Blk: 73216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 73216 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1ePlkOMYOjA6Sj3U5aK9R-vdkoYkQzZKOf49wW2bLE70= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 2 + LVPB: 1 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 35840 + LdsInitCVgprs: false + LdsNumBytes: 35840 + LdsNumElementsAlignedA: 2560 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2560 + LdsOffsetB_Blk: 68096 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 68096 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1R3pW5TGY8A5opfshY0CC3g5uD4DXiGRwS1VnQHSdtTs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 60672 + LdsInitCVgprs: false + LdsNumBytes: 60672 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 117760 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 60672 + LdsOffsetMetadata_Blk: 117760 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 2 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI13Sbf9PQEx5F-iKE9julI7_PdDt3yPapFP4vCFUpDo1E= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 62976 + LdsInitCVgprs: false + LdsNumBytes: 62976 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 62976 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 10 + NumLoadsB: 4 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI137t5Uvc0ssl50sE4SNlCcUp-XUoCcLX9_I4uGMXledk= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 54528 + LdsInitCVgprs: false + LdsNumBytes: 54528 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 54528 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 10 + NumLoadsB: 2 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIOmRHwGdcC8gFOr3dPzasFfQWs0PP2cDcwy6aKZU2f5U= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1454Bwxd_MIvQjuG-Y_CvHs_qnR7PgRRvHGhCUO-XbOY= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI1Rmqg4myY3vDAMf_l0hT_13f8mLa__KusFSpIvX00Gq4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42240 + LdsInitCVgprs: false + LdsNumBytes: 42240 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42240 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI15NEBMkxG3p9Ig0Cr9U7Vscm7bIb_z6qr2ylZCFd2y30= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16wxJkzA64nljgT6A4tMYPR-qnvL3NK7-LGLeTrL93K28= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 44544 + LdsInitCVgprs: false + LdsNumBytes: 44544 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 44544 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16s4XHH5Mm_pX3J5f5thK2zPFz4wk3wfVK1ZsVsKy_LpE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 36096 + LdsInitCVgprs: false + LdsNumBytes: 36096 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36096 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1XeU_kq-TCV7gFRA0zjadiPg5xVzK4J_9tcfhp6RHcE8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16uy_AuCnsXVT_W8ZUwPY2xZhbdvr0s1-k6m6FDS2GyPc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16JHWtGGP8udHdAtFC_ITt55YeNZjw7qy9YbaDQmBmLnw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25856 + LdsInitCVgprs: false + LdsNumBytes: 25856 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25856 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1CMycMeJchCKewZBbLPkfdDttQ7jpxPNi2DcgEi0aA7Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 1 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1536_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB3_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI167PfjxLOUHYcmsWqM3Jr5QHvep_HsB981kR15bVmyR_4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1536 + LdsBytesNoAmax: 34048 + LdsInitCVgprs: false + LdsNumBytes: 34048 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 24832 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34048 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsCoalescedB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16hW4G8jnTFUiWZWnW0Tmfmi9zpySUXkWHngzRijqsswA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 64 + LSPA: 16 + LSPB: 64 + LVCA: 16 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16pkwZPXEjbcu5SfItYZgOkhxquryDik7C3Mlw6DhdC9c= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 32 + LSPA: 16 + LSPB: 128 + LVCA: 16 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17664 + LdsInitCVgprs: false + LdsNumBytes: 17664 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 8448 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17664 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1ag0qOUjDoYcyoioR4OdNEh8L5HMCVTkf-1kVLG0Ke8Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 16 + LSPA: 16 + LSPB: 256 + LVCA: 16 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 71936 + LdsInitCVgprs: false + LdsNumBytes: 71936 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 4352 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71936 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1RElBQ1lA5tfxLjjMGZY4_t9WElkqvIrWw-ZvypHxsS4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 80384 + LdsInitCVgprs: false + LdsNumBytes: 80384 + LdsNumElementsAlignedA: 13824 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 13824 + LdsOffsetB_Blk: 144896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 80384 + LdsOffsetMetadata_Blk: 144896 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 16 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB4096_LBSPPM0_LPA32_LPB64_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1g4O7XQYq5Erjhq0mbnd0ZzsOn7Y74NZ0Ed50NCU8Cko= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 4096 + LdsBytesNoAmax: 71168 + LdsInitCVgprs: false + LdsNumBytes: 71168 + LdsNumElementsAlignedA: 4608 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4608 + LdsOffsetB_Blk: 135680 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71168 + LdsOffsetMetadata_Blk: 135680 + LdsPadA: 32 + LdsPadB: 64 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 64 + LSPA: 8 + LSPB: 64 + LVCA: 32 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16c-3ymAqE5iIzAVIl4M4mFas_yKBzwB8S5hIJHrx_G10= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 32 + LSPA: 8 + LSPB: 128 + LVCA: 32 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16l2koiE9edxgF5EcPYxeXQhdktTUf3O_Nh5I8-rOv_5c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16PDTjUmNdBPeXIoutsSh54JIRkDgOgTuGJ5_UUnXYC30= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16FZJXxZzBUYpkHY1uHTOo5xfvmHSftJ7Ss8JFmsnHOq4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 16 + LSPA: 8 + LSPB: 256 + LVCA: 32 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16UExxzleg69jKCOIPv3aHKRuxpeCJ0OtN8U8_i8380wE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 32 + LSPA: 8 + LSPB: 128 + LVCA: 32 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 25600 + LdsInitCVgprs: false + LdsNumBytes: 25600 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16486QKgKm2Q-ueCL9nw3616pMoSPYJwy96wpaKy2ZR9g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 64 + LSPA: 8 + LSPB: 64 + LVCA: 32 + LVCB: 4 + LVPA: 1 + LVPB: 4 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 41984 + LdsInitCVgprs: false + LdsNumBytes: 41984 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 41984 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1gJ-iaCReTZwgtOaq-r_7Ae1Xx5lpHHILXpRm4hXIFWo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34304 + LdsInitCVgprs: false + LdsNumBytes: 34304 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34304 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI1TOymj1f06yNxQYQNnDElBwV9Je3kDkNSQdL4ZFlGHxw= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1eVyecsmYTDI_tnvIi0T4aak3HoYfyCs1_BMjsWhSVGA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 16 + LSPA: 4 + LSPB: 256 + LVCA: 64 + LVCB: 1 + LVPA: 1 + LVPB: 16 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 84992 + LdsInitCVgprs: false + LdsNumBytes: 84992 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84992 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 4 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1SvhFCnFpOT5VXBaOtUT022uRF_6hTz1blD7UqHf2kpA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: false + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 32 + LSPA: 4 + LSPB: 128 + LVCA: 64 + LVCB: 2 + LVPA: 1 + LVPB: 8 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA1_GRVWB1_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA8_LPB16_LPM0_LRVW8_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU32_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM8_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bjlk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x32_MI16xDWExf5YqS6e0I09q-b7LtSPBEclPiz1pvMgNxDpD5yY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 32 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 1 + GlobalReadVectorWidthB: 1 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 32 + LSCB: 32 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 8 + LVPB: 8 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 6464 + LdsInitCVgprs: false + LdsNumBytes: 6464 + LdsNumElementsAlignedA: 1280 + LdsNumElementsAlignedB: 1280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 4096 + LdsOffsetB: 1280 + LdsOffsetB_Blk: 5376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 1280 + LdsOffsetMetadata_Blk: 5376 + LdsPadA: 8 + LdsPadB: 16 + LocalReadVectorWidth: 8 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 32 + MFMA_BF16_1K: false + MIBlock: [16, 16, 32, 1, 1, 1] + MIInputPerThread: 8 + MIInputPerThreadA: 8 + MIInputPerThreadB: 8 + MIInputPerThreadMetadata: 8 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 32 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 32, 1] + MaxLDS: 163840 + MbskPrefetchOpt: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: false + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: false + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 32 + _DepthUA: 32 + _DepthUB: 32 + _DepthUMetadata: 32 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 1 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 3 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction diff --git a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml index 497b1f0b34bc..5985f8a50ec4 100644 --- a/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml +++ b/projects/hipblaslt/library/src/amd_detail/rocblaslt/src/Tensile/Logic/asm_full/gfx950/gfx950/Origami/gfx950_Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs.yaml @@ -1,8 +1,10 @@ -- {MinimumRequiredVersion: 5.0.0} -- gfx950 -- gfx950 -- [Device 75a0] -- Activation: True +MinimumRequiredVersion: 5.0.0 +ScheduleName: gfx950 +ArchitectureName: gfx950 +CUCount: null +DeviceNames: [Device 75a0] +ProblemType: + Activation: true ActivationComputeDataType: 0 ActivationFuncCall: false ActivationNoGuard: false @@ -10,7 +12,7 @@ AllowNoFreeDims: false AssignedDerivedParameters: true Batched: true - BetaOnlyUseBias: True + BetaOnlyUseBias: true BiasDataTypeList: [0, 4] BiasSrc: D ComplexConjugateA: false @@ -79,31283 +81,22243 @@ UseScaleAB: Scalar UseScaleAlphaVec: 1 UseScaleCD: false -- - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MIrZs5mdCNmpikVNpV2YQQNFBCAdivVS8D1AsBXrhUP4k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 8] - MIWaveTileA: 8 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 256 - MacroTileA: 256 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 256 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 0 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 8 - ThreadTileA: 32 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MIADOZHpI0e4kz3PekZsq-nbyBZU7EZIxEg7KcvUC6g3g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 7] - MIWaveTileA: 8 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 224 - MacroTileA: 256 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 224 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 1 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 7 - ThreadTileA: 32 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIszVB7i9sCthoswarzra0qLsMXdGUHQA9HKGzu-1z5Kk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 6] - MIWaveTileA: 8 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 192 - MacroTileA: 256 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 2 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 6 - ThreadTileA: 32 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI4oZDINgtVU-f19tz2HYWh_LmdGFiZplMuFYBiLKIRVE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 124928 - LdsInitCVgprs: false - LdsNumBytes: 124928 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 5] - MIWaveTileA: 8 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 160 - MacroTileA: 256 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 3 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 5 - ThreadTileA: 32 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIQveu_FRPQ7kB2acKv85tmm0lWhIAYF1e-bVgSb8HhRU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 4] - MIWaveTileA: 8 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 128 - MacroTileA: 256 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 4 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 4 - ThreadTileA: 32 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1nNTR_ZcKmsj4A34uOMZTZX_oilYaQuaQFL2SNhs8cXo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 49152 - LdsInitCVgprs: false - LdsNumBytes: 49152 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 49152 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 3] - MIWaveTileA: 8 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 96 - MacroTileA: 256 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 5 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 3 - ThreadTileA: 32 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI17E3Elp9R2iGRIjoSUJt7ENU0IkIEDCYD27G5goO_zfQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 2] - MIWaveTileA: 8 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 64 - MacroTileA: 256 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 6 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 2 - ThreadTileA: 32 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI14oL9i2wAQa7K6yEREmecBQB_z40_9_5z5qamIYTi9BA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38912 - LdsInitCVgprs: false - LdsNumBytes: 38912 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38912 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 7 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MIVqWf-zHb4fGrqaI_PNfaMIuBSbf24WaOtDVKEmRwTy0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 166912 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 166912 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 8] - MIWaveTileA: 7 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 256 - MacroTileA: 224 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 224 - NumGlobalWriteVectorsPerThread: 224 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 8 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 8 - ThreadTileA: 28 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI2FE1_xdi6gisBeCfrvgksfzzbx9Yx6-LxacJqFoorOs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 71680 - LdsInitCVgprs: false - LdsNumBytes: 71680 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 166912 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 71680 - LdsOffsetMetadata_Blk: 166912 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 7] - MIWaveTileA: 7 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 224 - MacroTileA: 224 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 196 - NumGlobalWriteVectorsPerThread: 196 - NumLoadsA: 7 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 9 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 7 - ThreadTileA: 28 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MItgCKUscKBBiMtBOmdwUt6qAHeM_MR7YbKWEkh39p8UQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 129024 - LdsInitCVgprs: false - LdsNumBytes: 129024 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 6] - MIWaveTileA: 7 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 192 - MacroTileA: 224 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 168 - NumGlobalWriteVectorsPerThread: 168 - NumLoadsA: 7 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 10 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 6 - ThreadTileA: 28 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI_UOU6HZrGnyaniPidkPyA2vc3eMgOLn6XHhdMyr3gok= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 35840 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 5] - MIWaveTileA: 7 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 160 - MacroTileA: 224 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 140 - NumGlobalWriteVectorsPerThread: 140 - NumLoadsA: 7 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 11 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 5 - ThreadTileA: 28 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIdhuZzZrmslv6fIdikHYnhn1dpXNXnkfYYXLJoFZzt6s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 128 - MacroTileA: 224 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 12 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1b5RoEaxogDkty9pH0a4h4iqEK74VHsxBERA-H-NeUNY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 3] - MIWaveTileA: 7 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 96 - MacroTileA: 224 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 7 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 13 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 3 - ThreadTileA: 28 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1g3lxPhwHBbhj-5_vMz-ezv_Gct2IZefQGw3UsfUuGuM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 7 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 14 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1pRHgvlAR7N2hth7GOj12UHkPW1O97Tnl14UAnpJSv9c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 35840 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 35840 - LdsOffsetB_Blk: 101376 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 101376 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 7 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 15 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICt7le9q16ngZmPo_ZNYkoWOjOg2jVjonbKgLeTEzLxQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 8] - MIWaveTileA: 6 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 256 - MacroTileA: 192 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 192 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 16 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 8 - ThreadTileA: 24 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MIFCEugnbAE1o_KNOxQTjmgk1-riRp7-bqr1XYLHQkDJI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 129024 - LdsInitCVgprs: false - LdsNumBytes: 129024 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 7] - MIWaveTileA: 6 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 224 - MacroTileA: 192 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 168 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 6 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 17 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 7 - ThreadTileA: 24 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MIsXAzSnuik23-u_Tqj9IA9yIVlTTHvUnKVAh5tdPfE-0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 120832 - LdsInitCVgprs: false - LdsNumBytes: 120832 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 27648 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 6] - MIWaveTileA: 6 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 192 - MacroTileA: 192 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 6 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 18 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 6 - ThreadTileA: 24 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIdawCsxOV8Q1yfZkDghTj5ROObRZEHHk5r8TLSJ8-4DI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 5] - MIWaveTileA: 6 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 160 - MacroTileA: 192 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 19 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 5 - ThreadTileA: 24 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI96PVzcjbGKsmGAPHOizb5QovZZPMnBXqYWXYWCMR6RE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 4] - MIWaveTileA: 6 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 128 - MacroTileA: 192 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 20 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 4 - ThreadTileA: 24 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1XF9bGkyS2smPrKpZKRvXqZE_IlG80gFUg26F5mTZLis= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 21 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1cyGRGs5fNOjzNULtz2NEdDFcpvgD-53w5wck9Mz0rEA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 22 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1dXOVrMlvBwO6y7ESILkUB5p9GtB-Km9ykpVzg9YqhsY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 60416 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 60416 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 23 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI_FMKfIaIlw04mnTidFqT-Fnj0gDbPBKMLBy9AxGoipc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 124928 - LdsInitCVgprs: false - LdsNumBytes: 124928 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 8] - MIWaveTileA: 5 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 256 - MacroTileA: 160 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 160 - NumGlobalWriteVectorsPerThread: 160 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 24 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 8 - ThreadTileA: 20 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MIgvcBHckHBXx7rYEXvd58PMUeCwZSGsDJtMgKSLGOngQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 126976 - LdsInitCVgprs: false - LdsNumBytes: 126976 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 25600 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 7] - MIWaveTileA: 5 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 224 - MacroTileA: 160 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 140 - NumGlobalWriteVectorsPerThread: 140 - NumLoadsA: 5 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 25 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 7 - ThreadTileA: 20 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIiC0cBFJQlkymU09JdEq7dUq1wvcSX9wghKOSQxkOErI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 6] - MIWaveTileA: 5 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 192 - MacroTileA: 160 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 120 - NumGlobalWriteVectorsPerThread: 120 - NumLoadsA: 5 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 26 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 6 - ThreadTileA: 20 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MIwREMu9dXpHu0hNi1Te8Ryumj0HYpn1BFkusQoNfhAQ8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 5] - MIWaveTileA: 5 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 160 - MacroTileA: 160 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 100 - NumGlobalWriteVectorsPerThread: 100 - NumLoadsA: 5 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 27 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 5 - ThreadTileA: 20 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIcrglZoxFbHHjwhvMHFjwAoFdYaccIqWMvEsbqy0yalI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 28 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1BUsuevlFp3qE9qBkCpl8ZHtyZp2nTMvZX1nz1cQRT7M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 5 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 29 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1IVM_kq6fQOztyL87pw0KKbmpiglUkaPXmejIGutUrh4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 91136 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 91136 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 5 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 30 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1bgH1ukcWIV-QEjZUrEMgbOOF-CIy8Yu_mpQb3L9Zq3g= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 25600 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 25600 - LdsOffsetB_Blk: 58368 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 58368 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 5 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 31 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51712 - LdsInitCVgprs: false - LdsNumBytes: 51712 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51712 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 16 - LdsPadB: 16 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [32, 32, 64, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 256 - MacroTileA: 128 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 64 - MatrixInstM: 32 - MatrixInstN: 32 - MatrixInstruction: [32, 32, 64, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 4 - NonTemporalD: 4 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 16 - NumElementsPerThread: 128 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 32 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU8_SUM1_SUS256_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 8 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: true - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 4 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 2 - SubGroup1: 128 - SubGroupA: 2 - SubGroupB: 128 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 64 - ThreadTile1: 2 - ThreadTileA: 64 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 1 - enableLDSTrA: 0 - enableLDSTrB: 0 - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIHAxeqAl5GZyYpWyj9YBlHLiAot0B1CVx9Nf_5OHRC1Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 53248 - LdsInitCVgprs: false - LdsNumBytes: 53248 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 53248 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 224 - MacroTileA: 128 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 33 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIIlrnP4n1WnSi3fe0xoxepRfTuuLMMp-jgbnYdGnnvYE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 6] - MIWaveTileA: 4 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 192 - MacroTileA: 128 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 34 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 6 - ThreadTileA: 16 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI5bZ7yznefmkxpzzbB-XiEKlKRJu2AMeEgvXx60Tq5L0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 35 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI72iEPgzvyM2ZrWa9mJE7BjRqJtpIrgM81I_tGNEjK0c= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 36 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1xf_ukgQ-Z-qNS4jmkabhCAr3mB7WAQFBVo0jxB_s6wM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 37 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1j8uVA7i_pGx8ZCBrMR8ZhK3U6bp7J7YjtbYnnKdPxDI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 38 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1M6uR2wC2h6sr2Hl9YrF8zhWGQ7RgoVsQ-TM7lCE6pPI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22528 - LdsInitCVgprs: false - LdsNumBytes: 22528 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22528 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 39 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1lw4esJJcKXjrN7F_gKTEASvr4FPioQKVPOZImphrvBY= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 49152 - LdsInitCVgprs: false - LdsNumBytes: 49152 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 49152 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 8] - MIWaveTileA: 3 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 256 - MacroTileA: 96 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 96 - NumGlobalWriteVectorsPerThread: 96 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 40 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 8 - ThreadTileA: 12 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1JixbiZckG_HlXVPjjWZHNInPUwLf2xyuU7BTFa5yHI0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 7] - MIWaveTileA: 3 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 224 - MacroTileA: 96 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 84 - NumGlobalWriteVectorsPerThread: 84 - NumLoadsA: 3 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 41 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 7 - ThreadTileA: 12 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1wIqbeUYowBvH47eY0D90voHkVaoBg2lPdbX2d68nIJ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 3 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 42 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1J-2M-3q4LSCntZjUh3v8nZH3qUcDC89rCJ0Fdq3iMeI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 80896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 80896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 3 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 43 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1Vr1CG6o8iRf8BbhZJxSu0FVBXoX31LtlPuM8cs_YEXU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 44 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16Cnahg_B3KmoGxZkTO1OYZLBTHOrwHNY1VQd2t5IrylI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 3 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 45 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16W4yMmcRT_Rcp3ZNo2gSW3q4SJdd9i9DRRKcgqox59VE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24576 - LdsInitCVgprs: false - LdsNumBytes: 24576 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 24576 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 3 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 46 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16zNms4_iu6xgFQiq42WUbXrNnFNyslp04sbAKWAlTbog= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20480 - LdsInitCVgprs: false - LdsNumBytes: 20480 - LdsNumElementsAlignedA: 15360 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 15360 - LdsOffsetB_Blk: 48128 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20480 - LdsOffsetMetadata_Blk: 48128 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 3 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 47 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI11aFcbk23grO4-hK8aBkA0Q2TJ_RzSnSefhCsshDFWE8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 8] - MIWaveTileA: 2 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 256 - MacroTileA: 64 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 48 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 8 - ThreadTileA: 8 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1BIQ9uHrC9E8FHJ9mzReo0log_q-mFe47Gvfw9lRTxKU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 49 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1oWIDEj_bU2sIdW6WZSqKJI8Okx_mH0MwWvgKg4p0Wh0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 50 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1DfS6ucItLGFYy3q0Zz-JZd7icG_ycHgazunawsT8Pns= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 51 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1jK-MUAAqzxLFJYDxWtp_lRnOH6G_Sa3LS7gsDkzJ5ZI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 52 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16qBSTIiXAL1b2k0UyETAi0pu4GSfWwKXX5o-ybzBtncE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 24576 - LdsInitCVgprs: false - LdsNumBytes: 24576 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 24576 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 53 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16mOGETtdY_W1srxs4Mfgd04W-cV79aCR6raMD3HA6jWU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 18432 - LdsInitCVgprs: false - LdsNumBytes: 18432 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 18432 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 54 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16wIclDOa00a_BcP3sOUDpDc_qFY9RvbDE777DV1a9O7k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 25600 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 9216 - LdsOffsetMetadata_Blk: 25600 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 55 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1pg1aJvOuqnmTxe_-2e9_1S4uFGoAaPhgR-zs28ui024= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 38912 - LdsInitCVgprs: false - LdsNumBytes: 38912 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 38912 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 56 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1bJJ5iziycwP81Vg_rFbnRobq-WE2Gw-XuD7kvGV58LU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 40960 - LdsInitCVgprs: false - LdsNumBytes: 40960 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 35840 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 70656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 40960 - LdsOffsetMetadata_Blk: 70656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 1 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 57 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1c60kFRwy-kwCbpVB3Be-Tp-X0igfgGU9wMtdCOAL5xM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 32768 - LdsInitCVgprs: false - LdsNumBytes: 32768 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 32768 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 1 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 58 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI13_eugo-sWMKarMTBe1WI4hmzJdiGvToykcHBbj4WXow= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 25600 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 30720 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 1 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 59 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1gSIhM4a60TQQmpGca3zPH_1C9YMOrBSb2Bu9JrxvD-k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 22528 - LdsInitCVgprs: false - LdsNumBytes: 22528 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 22528 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 60 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16YH7g6N9l5w5imlP3Lp9mdDXwtO_ZtzNO9yUXMFVOjKQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 20480 - LdsInitCVgprs: false - LdsNumBytes: 20480 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 15360 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 37888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 20480 - LdsOffsetMetadata_Blk: 37888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 1 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 61 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16IFZMDGDs3-q6koIU2xIB_XUAfzhmng9Mg_46GO3Rqf4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 30720 - LdsInitCVgprs: false - LdsNumBytes: 30720 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 1 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 62 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16rLKaVqye5ZVlglTUO9ioHrZ6WB-GJ_Qs2RqP4S8ZIXE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 32 - LVCA: 8 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 5120 - LdsNumElementsAlignedB: 5120 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 16384 - LdsOffsetB: 5120 - LdsOffsetB_Blk: 21504 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 5120 - LdsOffsetMetadata_Blk: 21504 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 1 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 63 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MIyoGz0G9DWmdNz6WjS2Bq6rCsCrR_mXM4td8eMkMYEDw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73216 - LdsInitCVgprs: false - LdsNumBytes: 73216 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 38400 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 165888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73216 - LdsOffsetMetadata_Blk: 165888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 15] - MIWaveTileA: 4 - MIWaveTileB: 15 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 240 - MacroTileA: 256 - MacroTileB: 240 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 240 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 8 - NumLoadsB: 15 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 15 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 64 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 15 - ThreadTileA: 16 - ThreadTileB: 15 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIBnc7kWhy6AvjwbNc5adjziCfJiLAVAD1RzoNcS6OMZQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 68096 - LdsInitCVgprs: false - LdsNumBytes: 68096 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 33280 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 165888 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 68096 - LdsOffsetMetadata_Blk: 165888 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 13] - MIWaveTileA: 4 - MIWaveTileB: 13 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 208 - MacroTileA: 256 - MacroTileB: 208 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 52 - NumLoadsA: 8 - NumLoadsB: 13 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 13 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 65 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 13 - ThreadTileA: 16 - ThreadTileB: 13 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIF5A-1D1ZIPjJsJJ7ttt4AlfEFPQG0riur7Zv_qXG5YI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 128512 - LdsInitCVgprs: false - LdsNumBytes: 128512 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 28160 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 11] - MIWaveTileA: 4 - MIWaveTileB: 11 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 176 - MacroTileA: 256 - MacroTileB: 176 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 44 - NumLoadsA: 8 - NumLoadsB: 11 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 11 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 66 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 11 - ThreadTileA: 16 - ThreadTileB: 11 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIyRCs1CYgk6qoLPVzZz2698DWR4wMJ7whNVJ8JwJfnZc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123392 - LdsInitCVgprs: false - LdsNumBytes: 123392 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 9] - MIWaveTileA: 4 - MIWaveTileB: 9 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 144 - MacroTileA: 256 - MacroTileB: 144 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 8 - NumLoadsB: 9 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 9 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 67 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 9 - ThreadTileA: 16 - ThreadTileB: 9 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIaVDafPaNDXaLtu56b1YkZsQUI1aLmxb3jgdu_dv65EA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 17920 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 7] - MIWaveTileA: 4 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 112 - MacroTileA: 256 - MacroTileB: 112 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 8 - NumLoadsB: 7 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 7 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 68 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 7 - ThreadTileA: 16 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1jNVYmoC6fjSjbIADTf0f9l0Cs4TKifqjd6Xql0ehllk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 47616 - LdsInitCVgprs: false - LdsNumBytes: 47616 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 12800 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 47616 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 69 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1eX_SljyDEcS5rCPMhk2caJwb3hhe7qqzmb_HRJ0i-U0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 7680 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 70 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1fPZ5I9MI2WZKscvBb8L9C1UGdH6oze92UNxsd_rmjZ0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 8 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 32 - LSPB: 16 - LVCA: 8 - LVCB: 16 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 128 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37376 - LdsInitCVgprs: false - LdsNumBytes: 37376 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 2560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37376 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 71 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI17whl_3k4Y1EhiqZzn7y-o-UTqlat-Sd3AsiBr6V3nc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73216 - LdsInitCVgprs: false - LdsNumBytes: 73216 - LdsNumElementsAlignedA: 38400 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 38400 - LdsOffsetB_Blk: 169472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73216 - LdsOffsetMetadata_Blk: 169472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [15, 4] - MIWaveTileA: 15 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 240 - MacroTile1: 256 - MacroTileA: 240 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 240 - NumGlobalWriteVectorsPerThread: 240 - NumLoadsA: 15 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 15 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 72 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 60 - ThreadTile1: 4 - ThreadTileA: 60 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIjK835T0h-3Ar1N2b4kmmt65KtYLTJhfaZH3LmR2auFA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 68096 - LdsInitCVgprs: false - LdsNumBytes: 68096 - LdsNumElementsAlignedA: 33280 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33280 - LdsOffsetB_Blk: 164352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 68096 - LdsOffsetMetadata_Blk: 164352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [13, 4] - MIWaveTileA: 13 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 208 - MacroTile1: 256 - MacroTileA: 208 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 208 - NumGlobalWriteVectorsPerThread: 208 - NumLoadsA: 13 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 13 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 73 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 52 - ThreadTile1: 4 - ThreadTileA: 52 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIgsJynaoGp5Lr3OC9AUIiNlgAW9kkq7qPUmgK4ohOA_k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 128512 - LdsInitCVgprs: false - LdsNumBytes: 128512 - LdsNumElementsAlignedA: 28160 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 28160 - LdsOffsetB_Blk: 93696 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 28160 - LdsOffsetMetadata_Blk: 93696 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [11, 4] - MIWaveTileA: 11 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 176 - MacroTile1: 256 - MacroTileA: 176 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 176 - NumGlobalWriteVectorsPerThread: 176 - NumLoadsA: 11 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 11 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 74 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 44 - ThreadTile1: 4 - ThreadTileA: 44 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 0 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI6ZUiH-qsm4LTRuwyb7uw6oTXUaPI3Vjpt7WHaUbBOfI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 123392 - LdsInitCVgprs: false - LdsNumBytes: 123392 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 88576 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 23040 - LdsOffsetMetadata_Blk: 88576 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [9, 4] - MIWaveTileA: 9 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 144 - MacroTile1: 256 - MacroTileA: 144 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 144 - NumGlobalWriteVectorsPerThread: 144 - NumLoadsA: 9 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 9 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 75 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 36 - ThreadTile1: 4 - ThreadTileA: 36 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIiu6Dap7c-mOezayg2ePWEWJWIUNWwq_GUAWuc6asW8k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 52736 - LdsInitCVgprs: false - LdsNumBytes: 52736 - LdsNumElementsAlignedA: 17920 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17920 - LdsOffsetB_Blk: 83456 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 52736 - LdsOffsetMetadata_Blk: 83456 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [7, 4] - MIWaveTileA: 7 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 112 - MacroTile1: 256 - MacroTileA: 112 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 112 - NumGlobalWriteVectorsPerThread: 112 - NumLoadsA: 7 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 7 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 76 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 4 - ThreadTileA: 28 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1fyJI8fnAEv3sOPHcVPqOC2AU9kSqOftCPxastr6HZJE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 47616 - LdsInitCVgprs: false - LdsNumBytes: 47616 - LdsNumElementsAlignedA: 12800 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 12800 - LdsOffsetB_Blk: 78336 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 47616 - LdsOffsetMetadata_Blk: 78336 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 77 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-Zo3r6x3T98ZMWTZ4H7LqLtk7xXmH_6zOjhPN09FdTk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 42496 - LdsInitCVgprs: false - LdsNumBytes: 42496 - LdsNumElementsAlignedA: 7680 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 7680 - LdsOffsetB_Blk: 73216 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 42496 - LdsOffsetMetadata_Blk: 73216 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 78 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1wehOWUghk2A68IMMHO37hmuhc-kFrA6X9yfb9WRqhBg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 128 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 8 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 128 - LSCB: 128 - LSPA: 16 - LSPB: 32 - LVCA: 16 - LVCB: 8 - LVPA: 2 - LVPB: 2 - LdsBlockSizePerPadA: 128 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 37376 - LdsInitCVgprs: false - LdsNumBytes: 37376 - LdsNumElementsAlignedA: 2560 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 2560 - LdsOffsetB_Blk: 68096 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 37376 - LdsOffsetMetadata_Blk: 68096 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 79 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 128 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 128 - _DepthUA: 128 - _DepthUB: 128 - _DepthUMetadata: 128 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI1vCzuwEprjPexlyglCW3zRXsjX9xUmGeZzokhH60Z9yE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 8 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 2048 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 75776 - LdsInitCVgprs: false - LdsNumBytes: 75776 - LdsNumElementsAlignedA: 66560 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 66560 - LdsOffsetB_Blk: 197632 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 75776 - LdsOffsetMetadata_Blk: 197632 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [8, 1] - MIWaveTileA: 8 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 32 - MacroTileA: 256 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 80 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 8 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 32 - ThreadTile1: 1 - ThreadTileA: 32 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 8 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI1SiLGOArtDjYGdvuIvIxglRBR2PASF7znHipyq968J9k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81920 - LdsInitCVgprs: false - LdsNumBytes: 81920 - LdsNumElementsAlignedA: 64512 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 64512 - LdsOffsetB_Blk: 195584 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81920 - LdsOffsetMetadata_Blk: 195584 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 2] - MIWaveTileA: 7 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 64 - MacroTileA: 224 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 56 - NumLoadsA: 14 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 14 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 81 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 2 - ThreadTileA: 28 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI1Atkrqghs5hZ7rFrRDMn1qRHk9P7RjRIX2jTHJaCZcIs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 64512 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 64512 - LdsOffsetB_Blk: 195584 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 195584 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [7, 1] - MIWaveTileA: 7 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 224 - MacroTile1: 32 - MacroTileA: 224 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 14 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 14 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 82 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 28 - ThreadTile1: 1 - ThreadTileA: 28 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI1H6kBMzYhDsiXY1vBEU2Ud8UhLYehCYuWeQaCulrUTHo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 3] - MIWaveTileA: 6 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 96 - MacroTileA: 192 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 12 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 83 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 3 - ThreadTileA: 24 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI153vNbJPjGrVniCPDOzS5UGzSK6NOLhTT3dYrLoo1YJ4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 2] - MIWaveTileA: 6 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 64 - MacroTileA: 192 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 12 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 84 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 2 - ThreadTileA: 24 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1y0BaulKQmTtKr2XSp5P3dnaOK0k5UYUxfkhT1HxC_rU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 117760 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 117760 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [6, 1] - MIWaveTileA: 6 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 192 - MacroTile1: 32 - MacroTileA: 192 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 85 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 24 - ThreadTile1: 1 - ThreadTileA: 24 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MIUYoy0JGLYkrhIavz9gjog-Elk__Z0jYq-QXQEcN2PaA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 177152 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 177152 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 128 - MacroTileA: 160 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 10 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 86 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI1cUVPZBoZ-mtmE8sTWF_Az_FU-gB_6I8zhwfiN1pEaks= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 177152 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 177152 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 3] - MIWaveTileA: 5 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 96 - MacroTileA: 160 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 10 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 87 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 3 - ThreadTileA: 20 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI1EM3GcfaRjwnLLnXwjtnf8XZHB699buVut-TXk4ZjWyc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 63488 - LdsInitCVgprs: false - LdsNumBytes: 63488 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 63488 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 2] - MIWaveTileA: 5 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 64 - MacroTileA: 160 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 40 - NumLoadsA: 10 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 88 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 2 - ThreadTileA: 20 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI15N7cdBjXALKtpaYkT17ouFsMqBdqYm1rqUc2cC7GeLE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 46080 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 46080 - LdsOffsetB_Blk: 111616 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 111616 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [5, 1] - MIWaveTileA: 5 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 160 - MacroTile1: 32 - MacroTileA: 160 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 10 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 10 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 89 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 1 - ThreadTileA: 20 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MIU14cbwZ_t9NC-mBYr7N5MljJmeoGwBb-WUmjyy7ILhU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 160 - MacroTileA: 128 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 8 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 90 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIM08mmLtqT14x3TxGefabTcO290EciaeH6GnjDeU8Bgo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 4] - MIWaveTileA: 4 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 128 - MacroTileA: 128 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 64 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 91 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 4 - ThreadTileA: 16 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1OG3k0SIeaxxBYmDDXCQJ0v824MXDB2cvmAH14Io8mlM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 96 - MacroTileA: 128 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 8 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 92 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1Ee_Jq2fBrVEHXaPp-qikzlaFXpjSm2wKDPvD-MKxLM0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 2] - MIWaveTileA: 4 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 64 - MacroTileA: 128 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 93 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 2 - ThreadTileA: 16 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI13IeiqjCNSTMel5JBLjnHPiB9PE_JYWIIfXqePFminMc= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 128 - MacroTile1: 32 - MacroTileA: 128 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 94 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI1qUn-crPudL6dXZ6rCcKnkE9s4w-RcuNNqOhYdbIrR28= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 79872 - LdsInitCVgprs: false - LdsNumBytes: 79872 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 158720 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 79872 - LdsOffsetMetadata_Blk: 158720 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 6] - MIWaveTileA: 3 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 192 - MacroTileA: 96 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 72 - NumGlobalWriteVectorsPerThread: 72 - NumLoadsA: 6 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 95 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 6 - ThreadTileA: 12 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI18wp9zZc9oencMlWPaDY7BUc_iQgMxwF-RQOzHet7J3I= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 158720 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 158720 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 5] - MIWaveTileA: 3 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 160 - MacroTileA: 96 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 60 - NumGlobalWriteVectorsPerThread: 60 - NumLoadsA: 6 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 96 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 5 - ThreadTileA: 12 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI1LrtbILHycv46qRt6NxJYDFm4zT_v-o9V1TAKrHNeAeQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 128 - MacroTileA: 96 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 6 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 97 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI164Fx4NhdfXwLWoN6X-Afen5a5uy0a8LeKZgpjvani3CM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 3] - MIWaveTileA: 3 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 96 - MacroTileA: 96 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 36 - NumGlobalWriteVectorsPerThread: 36 - NumLoadsA: 6 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 98 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 3 - ThreadTileA: 12 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16_Nc2lY4fwUC-cbMivJ1U_eNf_9vhMW0WYu6jNMKBzYI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 2] - MIWaveTileA: 3 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 64 - MacroTileA: 96 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 6 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 99 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 2 - ThreadTileA: 12 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16X2E4Wx9NVLCyWp4uy3bGT7QFxR20eVc2o2HQJqwi8Hk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 27648 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 27648 - LdsOffsetB_Blk: 93184 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 93184 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 6 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 6 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 100 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI1ehl3l0UKCFUW0CqlYWmD5Gv6DpebCzV_7KPnbR_5y4Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81920 - LdsInitCVgprs: false - LdsNumBytes: 81920 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 64512 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81920 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 7] - MIWaveTileA: 2 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 224 - MacroTileA: 64 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 56 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 4 - NumLoadsB: 14 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 14 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 101 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 7 - ThreadTileA: 8 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI1PpwSxkO--tNKZnLh5u1Zqztz5GRs2qwDUtmWoQ32_0M= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 6] - MIWaveTileA: 2 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 192 - MacroTileA: 64 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 4 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 102 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 6 - ThreadTileA: 8 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1BIcxM2OYKG0Jqn5GRpKyHvCNhj2aQceoHMvTm7lJo6Q= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 63488 - LdsInitCVgprs: false - LdsNumBytes: 63488 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 63488 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 5] - MIWaveTileA: 2 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 160 - MacroTileA: 64 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 40 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 4 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 103 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 5 - ThreadTileA: 8 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1-eL7HedP-uoMH-4uLF-vatbGDwtmp3GZhw8IuLzDHvE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 4] - MIWaveTileA: 2 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 128 - MacroTileA: 64 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 104 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 4 - ThreadTileA: 8 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16Tidk1Ou4UA72T_vnHnWAsymx18z-7WG5S7yV6EMzxAE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 45056 - LdsInitCVgprs: false - LdsNumBytes: 45056 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 45056 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 3] - MIWaveTileA: 2 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 96 - MacroTileA: 64 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 105 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 3 - ThreadTileA: 8 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16P7nrnDgCqsPXJVFvUTHGSoy-i5STapWZDg4F8KHN2a8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 106 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI164FNCErQfeGGweRYwOlRLTlQQw-uihsMini7VG5MjqUs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 107 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI1Weg33rA_lR82ctIL5Wv-agG74LvDUGtagD8ZNXCm6m8= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 2048 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 75776 - LdsInitCVgprs: false - LdsNumBytes: 75776 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 66560 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 140288 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 75776 - LdsOffsetMetadata_Blk: 140288 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 8] - MIWaveTileA: 1 - MIWaveTileB: 8 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 256 - MacroTileA: 32 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 32 - NumGlobalWriteVectorsPerThread: 32 - NumLoadsA: 2 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 108 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 8 - ThreadTileA: 4 - ThreadTileB: 8 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 8 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI1yjWOfHPT7EqX9rhhXejY8_eiJ07brJZxCVBlCXR9DBo= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 73728 - LdsInitCVgprs: false - LdsNumBytes: 73728 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 64512 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 140288 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 73728 - LdsOffsetMetadata_Blk: 140288 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 7] - MIWaveTileA: 1 - MIWaveTileB: 7 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 224 - MacroTileA: 32 - MacroTileB: 224 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 28 - NumGlobalWriteVectorsPerThread: 28 - NumLoadsA: 2 - NumLoadsB: 14 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 14 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 109 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 7 - ThreadTileA: 4 - ThreadTileB: 7 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1pVP6pjfyPWPVeerMg8FY1EyiGjH4B_j7tc4lRUzRPvg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 61440 - LdsInitCVgprs: false - LdsNumBytes: 61440 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 61440 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 6] - MIWaveTileA: 1 - MIWaveTileB: 6 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 192 - MacroTileA: 32 - MacroTileB: 192 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 24 - NumGlobalWriteVectorsPerThread: 24 - NumLoadsA: 2 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 110 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 6 - ThreadTileA: 4 - ThreadTileB: 6 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI1q7mTED5UvVJrOPDvYFNPflDjUkX1okY5IJT3AZBddMM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 55296 - LdsInitCVgprs: false - LdsNumBytes: 55296 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 46080 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 55296 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 5] - MIWaveTileA: 1 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 160 - MacroTileA: 32 - MacroTileB: 160 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 20 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 2 - NumLoadsB: 10 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 10 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 111 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 5 - ThreadTileA: 4 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1fFpYWYjPWsmrBYgoz9CIhz7OHVoysPvt9SgTel6GPWI= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43008 - LdsInitCVgprs: false - LdsNumBytes: 43008 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43008 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 128 - MacroTileA: 32 - MacroTileB: 128 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 112 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16mu07e3PYZ_jp2LQxX0lmvu4pmsw_265fxGpIWycVSjg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 36864 - LdsInitCVgprs: false - LdsNumBytes: 36864 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 27648 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 74752 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 36864 - LdsOffsetMetadata_Blk: 74752 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 2 - NumLoadsB: 6 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 6 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 113 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI162x9lNzvJGF2jhOPNNRp1Q3mFaUkUkcNWlM7ecLmXXfs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26624 - LdsInitCVgprs: false - LdsNumBytes: 26624 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26624 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 114 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16-01AyvDsXfpTy-wvoK3K25-ZRKuFeCLIRsj5lVXzX_s= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 18432 - LdsInitCVgprs: false - LdsNumBytes: 18432 - LdsNumElementsAlignedA: 9216 - LdsNumElementsAlignedB: 9216 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 9216 - LdsOffsetB_Blk: 41984 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 18432 - LdsOffsetMetadata_Blk: 41984 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 115 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI1LDs52gy7NI4WuOMQ_idbNXBwBK86orsq1YVyOo4IvKs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 90624 - LdsInitCVgprs: false - LdsNumBytes: 90624 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 23040 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 90624 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 5] - MIWaveTileA: 4 - MIWaveTileB: 5 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 80 - MacroTileA: 256 - MacroTileB: 80 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 20 - NumLoadsA: 16 - NumLoadsB: 5 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 5 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 116 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 5 - ThreadTileA: 16 - ThreadTileB: 5 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1NztuuZzeaShCMYTqz64dRhvSvIEetpxjBMHS12EujLs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81408 - LdsInitCVgprs: false - LdsNumBytes: 81408 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 13824 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81408 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 3] - MIWaveTileA: 4 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 48 - MacroTileA: 256 - MacroTileB: 48 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 16 - NumLoadsB: 3 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 3 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 117 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 3 - ThreadTileA: 16 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1-eS3GREuZbaF_WcdTrJ8tJN1VRBcjitHDgD1jEhWCcE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 4 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 256 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 72192 - LdsInitCVgprs: false - LdsNumBytes: 72192 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 4608 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 72192 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [4, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 256 - MacroTile1: 16 - MacroTileA: 256 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 1 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 1 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 118 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 4 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 16 - SubGroup1: 16 - SubGroupA: 16 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 4 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [64, 4, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI1ZLUtUYYiOHV_3Bx1CweJ56xkOj29nbS2TUKPiIsgWeM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 90624 - LdsInitCVgprs: false - LdsNumBytes: 90624 - LdsNumElementsAlignedA: 23040 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 23040 - LdsOffsetB_Blk: 154112 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 90624 - LdsOffsetMetadata_Blk: 154112 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [5, 4] - MIWaveTileA: 5 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 80 - MacroTile1: 256 - MacroTileA: 80 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 80 - NumGlobalWriteVectorsPerThread: 80 - NumLoadsA: 5 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 5 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 119 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 20 - ThreadTile1: 4 - ThreadTileA: 20 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1QlOH2GDodFrn6A0rxCKHrFBaU_QWfv580nm9xWdZY0k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 81408 - LdsInitCVgprs: false - LdsNumBytes: 81408 - LdsNumElementsAlignedA: 13824 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 13824 - LdsOffsetB_Blk: 144896 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 81408 - LdsOffsetMetadata_Blk: 144896 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [3, 4] - MIWaveTileA: 3 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 48 - MacroTile1: 256 - MacroTileA: 48 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 48 - NumGlobalWriteVectorsPerThread: 48 - NumLoadsA: 3 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 3 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 120 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 4 - ThreadTileA: 12 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1PZ1TyY_JO23lBPKX9VhACTOX37NNgYiBjf2A-tanW7w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 256 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 - LDSTrInst: false - LSCA: 256 - LSCB: 256 - LSPA: 16 - LSPB: 16 - LVCA: 16 - LVCB: 16 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 256 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 72192 - LdsInitCVgprs: false - LdsNumBytes: 72192 - LdsNumElementsAlignedA: 4608 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 4608 - LdsOffsetB_Blk: 135680 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 72192 - LdsOffsetMetadata_Blk: 135680 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 4] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 256 - MacroTileA: 16 - MacroTileB: 256 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 16 - NumLoadsA: 1 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 1 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 121 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 256 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 64 - SubGroupA: 4 - SubGroupB: 64 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 4 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 16, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 256 - _DepthUA: 256 - _DepthUB: 256 - _DepthUMetadata: 256 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16Z6LYIKVcBKQJUXMfCTkH2UmyhUWm90vrVgFH5F5h_is= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 52224 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 52224 - LdsOffsetB_Blk: 183296 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 183296 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [3, 1] - MIWaveTileA: 3 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 96 - MacroTile1: 32 - MacroTileA: 96 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 12 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 12 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 122 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 12 - ThreadTile1: 1 - ThreadTileA: 12 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16Gx0c38MGhJUAjADEf6sgUUCYEnSxT0HvEevn-TLzDBA= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 2] - MIWaveTileA: 2 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 64 - MacroTileA: 64 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 16 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 123 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 2 - ThreadTileA: 8 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 2 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 32 - MacroTileA: 64 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 124 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 2 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 2 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16tO84-u8CmDozhzOYFPjZx1aAnLl_RdoGaIlGdosGUzM= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 69632 - LdsInitCVgprs: false - LdsNumBytes: 69632 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 52224 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 148480 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 69632 - LdsOffsetMetadata_Blk: 148480 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 3] - MIWaveTileA: 1 - MIWaveTileB: 3 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 96 - MacroTileA: 32 - MacroTileB: 96 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 12 - NumGlobalWriteVectorsPerThread: 12 - NumLoadsA: 4 - NumLoadsB: 12 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 12 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 125 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 3 - ThreadTileA: 4 - ThreadTileB: 3 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 51200 - LdsInitCVgprs: false - LdsNumBytes: 51200 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 51200 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 64 - MacroTileA: 32 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 8 - NumGlobalWriteVectorsPerThread: 8 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 126 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 2 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16nxXVLq-_LhX2N8-ggvlvGFb2CSQPdL0Jk31-3Yn8K8w= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 34816 - LdsInitCVgprs: false - LdsNumBytes: 34816 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 82944 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 34816 - LdsOffsetMetadata_Blk: 82944 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 4 - LoopUnroll: 512 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 127 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI1Uh3YqlpEBhxKyjcdbTJSpHtEarJLrJfEIa0Mnxwsb5k= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 67584 - LdsInitCVgprs: false - LdsNumBytes: 67584 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 164864 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 67584 - LdsOffsetMetadata_Blk: 164864 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 1 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 8 - LoopUnroll: 1024 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [2, 2] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 32 - MacroTileA: 32 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 128 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 8 - SubGroup1: 32 - SubGroupA: 8 - SubGroupB: 32 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [32, 8, 1] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16t0-mPljiEZtupRhWAFpnxBDrJen-FsvDD34Q7PN3s_4= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 17408 - LdsInitCVgprs: false - LdsNumBytes: 17408 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 17408 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 2 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 129 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16CAJPr2P6olRKJqA2Fmiymyoh6JhX7nYYGS7ffsGJvmw= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 17408 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 17408 - LdsOffsetB_Blk: 50176 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 50176 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 130 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16K_Na50sAdM91rhXevAlyiM9lrmobRF_PSYV6N76rQzk= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 34816 - LdsNumElementsAlignedB: 8704 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 34816 - LdsOffsetB_Blk: 100352 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 100352 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 8 - NumLoadsB: 2 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 2 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 131 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16dVDH5F1pzH8NV-uymOAiVxprnNui5yf2Mp5t1_LqLTs= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 26112 - LdsInitCVgprs: false - LdsNumBytes: 26112 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 17408 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 32768 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 41472 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 26112 - LdsOffsetMetadata_Blk: 41472 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 2 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 132 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16Uv-d7icfFxrZ-YUiRaTBsX3xXbc3HcmSMlEdtwh8Yb0= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 0 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 512 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 512 - LSCB: 512 - LSPA: 8 - LSPB: 8 - LVCA: 32 - LVCB: 32 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 512 - LdsBlockSizePerPadB: 512 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 43520 - LdsInitCVgprs: false - LdsNumBytes: 43520 - LdsNumElementsAlignedA: 8704 - LdsNumElementsAlignedB: 34816 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 8704 - LdsOffsetB_Blk: 74240 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 43520 - LdsOffsetMetadata_Blk: 74240 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 1 - LoopUnroll: 128 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 2 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 2 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 0 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 133 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 512 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 512 - _DepthUA: 512 - _DepthUB: 512 - _DepthUMetadata: 512 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1TNTLRLqQ0uhz1zq5XPt3rxttGE1MT5QoFsMWFmpwZGg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 33792 - LdsInitCVgprs: false - LdsNumBytes: 33792 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 33792 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 1] - MIWaveTileA: 1 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 16 - MacroTileA: 16 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 1 - NumGlobalWriteVectorsPerThread: 1 - NumLoadsA: 4 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 134 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 1 - ThreadTileA: 4 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI13HHr0kFgQAqJ2TjtZ8tTohe6Udz-5u0CwkMq-_9renQ= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 33792 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 33792 - LdsOffsetB_Blk: 99328 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 99328 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [2, 1] - MIWaveTileA: 2 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 32 - MacroTile1: 16 - MacroTileA: 32 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 8 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 8 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 135 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 8 - ThreadTile1: 1 - ThreadTileA: 8 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1KAql7WR7THsIAKFHzmlcHQVpFofe0Vzt2xFE0wiTOfU= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84480 - LdsInitCVgprs: false - LdsNumBytes: 84480 - LdsNumElementsAlignedA: 67584 - LdsNumElementsAlignedB: 16896 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 67584 - LdsOffsetB_Blk: 198656 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84480 - LdsOffsetMetadata_Blk: 198656 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [4, 1] - MIWaveTileA: 4 - MIWaveTileB: 1 - MIWaveTileMetadata: 0 - MacroTile0: 64 - MacroTile1: 16 - MacroTileA: 64 - MacroTileB: 16 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 16 - NumLoadsB: 4 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 16 - NumLoadsPerpendicularB: 4 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 136 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 16 - ThreadTile1: 1 - ThreadTileA: 16 - ThreadTileB: 1 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1-LGVgPAQHxTiHAOxIZA3A-LLqgvwFhsRG4LoerQvYqg= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 50688 - LdsInitCVgprs: false - LdsNumBytes: 50688 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 33792 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 65536 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 82432 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 50688 - LdsOffsetMetadata_Blk: 82432 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 2] - MIWaveTileA: 1 - MIWaveTileB: 2 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 32 - MacroTileA: 16 - MacroTileB: 32 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 2 - NumGlobalWriteVectorsPerThread: 2 - NumLoadsA: 4 - NumLoadsB: 8 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 8 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 137 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 2 - ThreadTileA: 4 - ThreadTileB: 2 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false - - 1LDSBuffer: 1 - ActivationAlt: false - ActivationFuncCall: false - ActivationFused: true - AssertAIGreaterThanEqual: -1 - AssertAILessThanEqual: -1 - AssertFree0ElementMultiple: 1 - AssertFree1ElementMultiple: 1 - AssertSummationElementMultiple: 1 - AssignedDerivedParameters: true - AssignedProblemIndependentDerivedParameters: true - BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1uwD3q8H7ROxJTYYpP8cvtJiz3DeoGvbO5IiJaZpCFnE= - BufferLoad: true - BufferStore: true - CUCount: null - CUOccupancy: -1 - ClusterLocalRead: 1 - CodeObjectVersion: '4' - ConvertAfterDS: false - CustomKernelName: '' - DebugStreamK: 0 - DepthU: 1024 - DirectToLds: 0 - DirectToLdsA: false - DirectToLdsB: false - DirectToVgprA: false - DirectToVgprB: false - DirectToVgprSparseMetadata: false - EdgeType: ShiftPtr - EnableF32XEmulationLds: false - EnableF32XdlMathOp: false - EnableMatrixInstruction: true - ExpandPointerSwap: false - ExpertSchedulingMode: 0 - ForceDisableShadowInit: false - GlobalReadPerMfma: 1.0 - GlobalReadVectorWidthA: 16 - GlobalReadVectorWidthB: 16 - GlobalSplitU: 0 - GlobalSplitUAlgorithm: MultipleBuffer - GlobalSplitUCoalesced: false - GlobalSplitUWorkGroupMappingRoundRobin: false - GlobalWriteVectorWidth: 1 - GroupLoadStore: false - GuaranteeNoPartialA: true - GuaranteeNoPartialB: true - GuaranteeNoPartialMetadata: true - ISA: [9, 5, 0] - InnerUnroll: 1 - InterleaveAlpha: 0 - InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, - SupportUserGSU: false, UseUniversalArgs: true} - Kernel: true - KernelLanguage: Assembly - KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 - LDSTrInst: false - LSCA: 1024 - LSCB: 1024 - LSPA: 4 - LSPB: 4 - LVCA: 64 - LVCB: 64 - LVPA: 1 - LVPB: 1 - LdsBlockSizePerPadA: 1024 - LdsBlockSizePerPadB: 1024 - LdsBlockSizePerPadMetadata: 0 - LdsBytesNoAmax: 84480 - LdsInitCVgprs: false - LdsNumBytes: 84480 - LdsNumElementsAlignedA: 16896 - LdsNumElementsAlignedB: 67584 - LdsNumElementsAlignedMetadata: 0 - LdsOffsetA: 0 - LdsOffsetA_Blk: 131072 - LdsOffsetB: 16896 - LdsOffsetB_Blk: 147968 - LdsOffsetBias: 0 - LdsOffsetBiasGSU: 0 - LdsOffsetBiasNonGSU: 0 - LdsOffsetMetadata: 84480 - LdsOffsetMetadata_Blk: 147968 - LdsPadA: 32 - LdsPadB: 32 - LdsPadMetadata: 0 - LocalReadVectorWidth: 16 - LocalSplitU: 4 - LocalSplitUReuseLDS: 1 - LocalWritePerMfma: -1 - LocalWriteUseSgprA: false - LocalWriteUseSgprB: false - LoopIters: 2 - LoopUnroll: 256 - MFMA_BF16_1K: false - MIArchVgpr: false - MIBlock: [16, 16, 128, 1, 1, 1] - MIInputPerThread: 32 - MIInputPerThreadA: 32 - MIInputPerThreadB: 32 - MIInputPerThreadMetadata: 32 - MIOutputVectorWidth: 4 - MIRegPerOut: 1 - MIWaveGroup: [1, 1] - MIWaveTile: [1, 4] - MIWaveTileA: 1 - MIWaveTileB: 4 - MIWaveTileMetadata: 0 - MacroTile0: 16 - MacroTile1: 64 - MacroTileA: 16 - MacroTileB: 64 - MagicDivAlg: 2 - MathClocksUnrolledLoop: 0 - MatrixInstB: 1 - MatrixInstBM: 1 - MatrixInstBN: 1 - MatrixInstK: 128 - MatrixInstM: 16 - MatrixInstN: 16 - MatrixInstruction: [16, 16, 128, 1] - MaxLDS: 163840 - MaxOccupancy: 40 - MbskPrefetchMethod: 0 - NoLdsWriteCode: false - NoReject: false - NoTailLoop: false - NonDTLTailLoopA: true - NonDTLTailLoopB: true - NonTemporal: -1 - NonTemporalA: 0 - NonTemporalB: 0 - NonTemporalC: 0 - NonTemporalD: 0 - NonTemporalE: 0 - NonTemporalMetadata: 0 - NonTemporalWS: 0 - NumElementsPerBatchStore: 0 - NumElementsPerThread: 4 - NumGlobalWriteVectorsPerThread: 4 - NumLoadsA: 4 - NumLoadsB: 16 - NumLoadsCoalescedA: 1 - NumLoadsCoalescedB: 1 - NumLoadsPerpendicularA: 4 - NumLoadsPerpendicularB: 16 - NumThreads: 256 - NumWaveSplitK: 1 - OptNoLoadLoop: 1 - PackedC0IdxChars: [I] - PackedC0IndicesX: [0] - PackedC1IdxChars: [J] - PackedC1IndicesX: [1] - PrefetchGlobalRead: 2 - PrefetchLocalRead: 1 - PreloadKernArgs: true - ScheduleGlobalRead: 1 - ScheduleIterAlg: 3 - ScheduleLocalWrite: 1 - SolutionIndex: 138 - SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 - SourceSwap: true - StaggerU: 0 - StaggerUMapping: 1 - StaggerUStride: 1024 - StorePriorityOpt: false - StoreRemapVectorWidth: 0 - StoreSwapAddr: false - StoreSyncOpt: 0 - StoreVectorWidth: 1 - StreamK: 5 - StreamKAtomic: 0 - StreamKXCCMapping: 0 - SubGroup0: 4 - SubGroup1: 16 - SubGroupA: 4 - SubGroupB: 16 - SuppressNoLoadLoop: false - ThreadTile: [1, 1] - ThreadTile0: 4 - ThreadTile1: 4 - ThreadTileA: 4 - ThreadTileB: 4 - TransposeLDS: 1 - TransposeLDSMetadata: 1 - ULSGRODoubleG2L: 0 - UnrollLoopSwapGlobalReadOrder: 0 - UnrollMajorLDSA: true - UnrollMajorLDSB: true - UnrollMajorLDSMetadata: true - Use64bShadowLimit: true - UseDotInstruction: false - UseF32XEmulation: false - UseInstOffsetForGRO: 0 - UseSgprForGRO: 0 - Valid: true - VectorStore: -1 - VectorWidthA: 1 - VectorWidthB: 1 - WaveSeparateGlobalReadA: 0 - WaveSeparateGlobalReadB: 0 - WaveSeparateGlobalReadMetadata: 0 - WaveSplitK: false - WavefrontSize: 64 - WorkGroup: [16, 4, 4] - WorkGroupMapping: 0 - WorkGroupMappingXCC: -1 - WorkGroupMappingXCCGroup: -1 - WorkGroupReduction: false - WorkspaceCheck: [4, 0, 0] - _DepthU: 1024 - _DepthUA: 1024 - _DepthUB: 1024 - _DepthUMetadata: 1024 - _GlobalAccumulation: PartialsBuffer - _UseSgprForGRO: 0 - _VectorStore: 1 - _WorkspaceSizePerElemBias: 0 - _WorkspaceSizePerElemC: 4 - _staggerStrideShift: 0 - enableLDSTrA: false - enableLDSTrB: false - reorderGRInstForDTVA: false - reorderGRInstForDTVB: false - tailLoopOptA: false - tailLoopOptB: false -- null -- null -- null -- null -- DeviceEfficiency -- Prediction +DefaultSolution: + 1LDSBuffer: 0 + ActivationAlt: false + ActivationFuncCall: true + ActivationFused: true + AdaptiveGemm: 0 + AdaptiveGemmGSUA: 0 + AdaptiveGemmNTAB: 0 + AssertAIGreaterThanEqual: -1 + AssertAILessThanEqual: -1 + AssertFree0ElementMultiple: 1 + AssertFree1ElementMultiple: 1 + AssertSummationElementMultiple: 1 + BufferLoad: true + BufferStore: true + ClusterDim: [1, 1] + ClusterLocalRead: 1 + CompactLoopStore: false + ConvertAfterDS: false + CustomKernelName: '' + DebugPersistentKernelLoopForever: false + DebugStreamK: 0 + DepthU: -1 + DirectToLds: 0 + DirectToLdsMetadata: 1 + DirectToVgprA: false + DirectToVgprB: false + DirectToVgprMXSA: false + DirectToVgprMXSB: false + DirectToVgprSparseMetadata: false + DtlPlusLdsBuf: -1 + ExpandPointerSwap: true + ExtraLatencyForLR: 0 + ExtraMiLatencyLeft: -1 + ForceDisableShadowInit: false + GlobalReadPerMfma: 1.0 + GlobalReadVectorWidthA: -1 + GlobalReadVectorWidthB: -1 + GlobalSplitU: 1 + GlobalSplitUAlgorithm: MultipleBuffer + GlobalSplitUCoalesced: false + GlobalSplitUWorkGroupMappingRoundRobin: false + GroupLoadStore: false + HalfPLR: 0 + InitCIterWmma: -1 + InnerUnroll: 1 + InterleaveAlpha: 0 + KernelLanguage: Assembly + LDSSegmentInterleave: 0 + LDSTrInst: false + LdsBlockSizePerPadA: -1 + LdsBlockSizePerPadB: -1 + LdsBlockSizePerPadMXSA: -1 + LdsBlockSizePerPadMXSB: -1 + LdsBlockSizePerPadMetadata: 0 + LdsPadA: -1 + LdsPadB: -1 + LdsPadMXSA: -1 + LdsPadMXSB: -1 + LdsPadMetadata: 0 + LocalReadVectorWidth: -1 + LocalReadVectorWidthA: -1 + LocalReadVectorWidthB: -1 + LocalWritePerMfma: -1 + MIArchVgpr: false + MXLoadInst: Auto + MXScaleFormat: Auto + MagicDivAlg: 2 + MatrixInstruction: [] + MaxLDS: -1 + MaxOccupancy: 40 + MbskPrefetchMethod: -1 + MinGRIncPerMfma: -1 + NoReject: false + NonTemporal: -1 + NonTemporalA: 0 + NonTemporalB: 0 + NonTemporalC: 0 + NonTemporalD: 0 + NonTemporalE: 0 + NonTemporalGate: 0 + NonTemporalMXSA: 0 + NonTemporalMXSB: 0 + NonTemporalMetadata: 0 + NonTemporalWS: 0 + NonVolatile: -1 + NonVolatileA: 0 + NonVolatileB: 0 + NonVolatileC: 0 + NonVolatileD: 0 + NonVolatileE: 0 + NonVolatileMXSA: 0 + NonVolatileMXSB: 0 + NonVolatileMetadata: 0 + NonVolatileWS: 0 + NumElementsPerBatchStore: 0 + NumLoadsCoalescedA: 1 + NumLoadsCoalescedB: 1 + OptNoLoadLoop: 1 + PrefetchAcrossPersistent: 0 + PrefetchGL2: 0 + PrefetchGlobalRead: 1 + PrefetchLocalRead: 1 + PreloadKernArgs: true + SFCWGM: + - [1, 1] + - [1, 1] + ScheduleGROverBarrier: -1 + ScheduleGlobalRead: 1 + ScheduleIterAlg: 3 + ScheduleLocalWrite: 1 + SourceSwap: false + SpaceFillingAlgo: [] + StaggerU: 32 + StaggerUMapping: 0 + StaggerUStride: 256 + StorePriorityOpt: false + StoreRemapVectorWidth: 0 + StoreSyncOpt: 0 + StoreVectorWidth: -1 + StreamK: 0 + StreamKAtomic: 0 + StreamKFixupTreeReduction: 0 + StreamKForceDPOnly: 0 + StreamKWorkStealing: 0 + StreamKXCCMapping: 0 + SuppressNoLoadLoop: false + SwInstructionPrefetch: 1 + SwapGlobalReadOrder: 0 + TDMInst: 0 + TDMIterateMode: 0 + TDMSplit: false + TailloopInNll: false + TemporalHint: -1 + TemporalHintA: 0 + TemporalHintB: 0 + TemporalHintC: 0 + TemporalHintD: 0 + TemporalHintE: 0 + TemporalHintMXSA: 0 + TemporalHintMXSB: 0 + TemporalHintMetadata: 0 + TemporalHintWS: 0 + ThreadTile: [4, 4] + TransposeLDS: -1 + TransposeLDSMetadata: -1 + UnrollLoopSwapGlobalReadOrder: 0 + Use64bShadowLimit: true + Use64bShadowLimitMX: false + UseCustomMainLoopSchedule: -1 + UseDualFMAC: false + UseInstOffsetForGRO: 0 + UsePLRPack: 0 + UseSgprForGRO: -1 + UseSubtileImpl: false + VectorStore: -1 + VectorWidthA: -1 + VectorWidthB: -1 + WaveSeparateGlobalReadA: 0 + WaveSeparateGlobalReadB: 0 + WaveSeparateGlobalReadMetadata: 0 + WaveSplitK: false + WavefrontSize: -1 + WorkGroup: [16, 16, 1] + WorkGroupMapping: 8 + WorkGroupMappingXCC: 1 + WorkGroupMappingXCCGroup: -1 + WorkGroupReduction: false +Solutions: +- SolutionIndex: 0 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x256x128_MIrZs5mdCNmpikVNpV2YQQNFBCAdivVS8D1AsBXrhUP4k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 8] + MIWaveTileA: 8 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 256 + MacroTileA: 256 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 256 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 8 + ThreadTileA: 32 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 1 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x224x128_MIADOZHpI0e4kz3PekZsq-nbyBZU7EZIxEg7KcvUC6g3g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 7] + MIWaveTileA: 8 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 224 + MacroTileA: 256 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 224 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 7 + ThreadTileA: 32 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 2 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x192x128_MIszVB7i9sCthoswarzra0qLsMXdGUHQA9HKGzu-1z5Kk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 6] + MIWaveTileA: 8 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 192 + MacroTileA: 256 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 6 + ThreadTileA: 32 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 3 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x160x128_MI4oZDINgtVU-f19tz2HYWh_LmdGFiZplMuFYBiLKIRVE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 124928 + LdsInitCVgprs: false + LdsNumBytes: 124928 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 5] + MIWaveTileA: 8 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 160 + MacroTileA: 256 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 5 + ThreadTileA: 32 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 4 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x128x128_MIQveu_FRPQ7kB2acKv85tmm0lWhIAYF1e-bVgSb8HhRU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 4] + MIWaveTileA: 8 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 128 + MacroTileA: 256 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 4 + ThreadTileA: 32 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 5 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x96x128_MI1nNTR_ZcKmsj4A34uOMZTZX_oilYaQuaQFL2SNhs8cXo= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 49152 + LdsInitCVgprs: false + LdsNumBytes: 49152 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 49152 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 3] + MIWaveTileA: 8 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 96 + MacroTileA: 256 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 3 + ThreadTileA: 32 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 6 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x64x128_MI17E3Elp9R2iGRIjoSUJt7ENU0IkIEDCYD27G5goO_zfQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 2] + MIWaveTileA: 8 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 64 + MacroTileA: 256 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 2 + ThreadTileA: 32 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 7 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x128_MI14oL9i2wAQa7K6yEREmecBQB_z40_9_5z5qamIYTi9BA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 38912 + LdsInitCVgprs: false + LdsNumBytes: 38912 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38912 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 8 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x256x128_MIVqWf-zHb4fGrqaI_PNfaMIuBSbf24WaOtDVKEmRwTy0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 166912 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 166912 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 8] + MIWaveTileA: 7 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 256 + MacroTileA: 224 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 224 + NumGlobalWriteVectorsPerThread: 224 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 8 + ThreadTileA: 28 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 9 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x224x128_MI2FE1_xdi6gisBeCfrvgksfzzbx9Yx6-LxacJqFoorOs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 71680 + LdsInitCVgprs: false + LdsNumBytes: 71680 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 166912 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 71680 + LdsOffsetMetadata_Blk: 166912 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 7] + MIWaveTileA: 7 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 224 + MacroTileA: 224 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 196 + NumGlobalWriteVectorsPerThread: 196 + NumLoadsA: 7 + NumLoadsB: 7 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 7 + ThreadTileA: 28 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 10 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x192x128_MItgCKUscKBBiMtBOmdwUt6qAHeM_MR7YbKWEkh39p8UQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 129024 + LdsInitCVgprs: false + LdsNumBytes: 129024 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 6] + MIWaveTileA: 7 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 192 + MacroTileA: 224 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 168 + NumGlobalWriteVectorsPerThread: 168 + NumLoadsA: 7 + NumLoadsB: 6 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 6 + ThreadTileA: 28 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 11 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x160x128_MI_UOU6HZrGnyaniPidkPyA2vc3eMgOLn6XHhdMyr3gok= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 35840 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 5] + MIWaveTileA: 7 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 160 + MacroTileA: 224 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 140 + NumGlobalWriteVectorsPerThread: 140 + NumLoadsA: 7 + NumLoadsB: 5 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 5 + ThreadTileA: 28 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 12 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x128x128_MIdhuZzZrmslv6fIdikHYnhn1dpXNXnkfYYXLJoFZzt6s= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 128 + MacroTileA: 224 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 4 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 13 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x96x128_MI1b5RoEaxogDkty9pH0a4h4iqEK74VHsxBERA-H-NeUNY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 3] + MIWaveTileA: 7 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 96 + MacroTileA: 224 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 7 + NumLoadsB: 3 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 3 + ThreadTileA: 28 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 14 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x128_MI1g3lxPhwHBbhj-5_vMz-ezv_Gct2IZefQGw3UsfUuGuM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 7 + NumLoadsB: 2 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 15 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x128_MI1pRHgvlAR7N2hth7GOj12UHkPW1O97Tnl14UAnpJSv9c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 35840 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 35840 + LdsOffsetB_Blk: 101376 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 101376 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 7 + NumLoadsB: 1 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 16 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x256x128_MICt7le9q16ngZmPo_ZNYkoWOjOg2jVjonbKgLeTEzLxQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 8] + MIWaveTileA: 6 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 256 + MacroTileA: 192 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 192 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 8 + ThreadTileA: 24 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 17 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x224x128_MIFCEugnbAE1o_KNOxQTjmgk1-riRp7-bqr1XYLHQkDJI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 129024 + LdsInitCVgprs: false + LdsNumBytes: 129024 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 7] + MIWaveTileA: 6 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 224 + MacroTileA: 192 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 168 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 6 + NumLoadsB: 7 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 7 + ThreadTileA: 24 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 18 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x192x128_MIsXAzSnuik23-u_Tqj9IA9yIVlTTHvUnKVAh5tdPfE-0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 120832 + LdsInitCVgprs: false + LdsNumBytes: 120832 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 27648 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 6] + MIWaveTileA: 6 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 192 + MacroTileA: 192 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 6 + NumLoadsB: 6 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 6 + ThreadTileA: 24 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 19 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x160x128_MIdawCsxOV8Q1yfZkDghTj5ROObRZEHHk5r8TLSJ8-4DI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 5] + MIWaveTileA: 6 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 160 + MacroTileA: 192 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 5 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 5 + ThreadTileA: 24 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 20 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x128x128_MI96PVzcjbGKsmGAPHOizb5QovZZPMnBXqYWXYWCMR6RE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 4] + MIWaveTileA: 6 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 128 + MacroTileA: 192 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 4 + ThreadTileA: 24 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 21 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x128_MI1XF9bGkyS2smPrKpZKRvXqZE_IlG80gFUg26F5mTZLis= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 3 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 22 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x128_MI1cyGRGs5fNOjzNULtz2NEdDFcpvgD-53w5wck9Mz0rEA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 23 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x128_MI1dXOVrMlvBwO6y7ESILkUB5p9GtB-Km9ykpVzg9YqhsY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 60416 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 60416 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 1 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 24 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x256x128_MI_FMKfIaIlw04mnTidFqT-Fnj0gDbPBKMLBy9AxGoipc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 124928 + LdsInitCVgprs: false + LdsNumBytes: 124928 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 8] + MIWaveTileA: 5 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 256 + MacroTileA: 160 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 160 + NumGlobalWriteVectorsPerThread: 160 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 8 + ThreadTileA: 20 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 25 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x224x128_MIgvcBHckHBXx7rYEXvd58PMUeCwZSGsDJtMgKSLGOngQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 126976 + LdsInitCVgprs: false + LdsNumBytes: 126976 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 25600 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 7] + MIWaveTileA: 5 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 224 + MacroTileA: 160 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 140 + NumGlobalWriteVectorsPerThread: 140 + NumLoadsA: 5 + NumLoadsB: 7 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 7 + ThreadTileA: 20 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 26 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x192x128_MIiC0cBFJQlkymU09JdEq7dUq1wvcSX9wghKOSQxkOErI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 6] + MIWaveTileA: 5 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 192 + MacroTileA: 160 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 120 + NumGlobalWriteVectorsPerThread: 120 + NumLoadsA: 5 + NumLoadsB: 6 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 6 + ThreadTileA: 20 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 27 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x160x128_MIwREMu9dXpHu0hNi1Te8Ryumj0HYpn1BFkusQoNfhAQ8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 5] + MIWaveTileA: 5 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 160 + MacroTileA: 160 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 100 + NumGlobalWriteVectorsPerThread: 100 + NumLoadsA: 5 + NumLoadsB: 5 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 5 + ThreadTileA: 20 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 28 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x128_MIcrglZoxFbHHjwhvMHFjwAoFdYaccIqWMvEsbqy0yalI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 4 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 29 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x128_MI1BUsuevlFp3qE9qBkCpl8ZHtyZp2nTMvZX1nz1cQRT7M= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 5 + NumLoadsB: 3 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 30 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x128_MI1IVM_kq6fQOztyL87pw0KKbmpiglUkaPXmejIGutUrh4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 91136 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 91136 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 5 + NumLoadsB: 2 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 31 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x128_MI1bgH1ukcWIV-QEjZUrEMgbOOF-CIy8Yu_mpQb3L9Zq3g= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 25600 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 25600 + LdsOffsetB_Blk: 58368 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 58368 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 5 + NumLoadsB: 1 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 32 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x256x128_MI32x32x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA16_LPB16_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC4_NTD4_NTM0_NEPBS16_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU8_SUM1_SUS256_SPO1_SRVW0_SSO4_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 51712 + LdsInitCVgprs: false + LdsNumBytes: 51712 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51712 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 16 + LdsPadB: 16 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [32, 32, 64, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 256 + MacroTileA: 128 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 64 + MatrixInstM: 32 + MatrixInstN: 32 + MatrixInstruction: [32, 32, 64, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NonTemporalC: 4 + NonTemporalD: 4 + NumElementsPerBatchStore: 16 + NumElementsPerThread: 128 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 8 + StaggerUMapping: 1 + StorePriorityOpt: true + StoreSwapAddr: false + StoreSyncOpt: 4 + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 2 + SubGroup1: 128 + SubGroupA: 2 + SubGroupB: 128 + ThreadTile: [1, 1] + ThreadTile0: 64 + ThreadTile1: 2 + ThreadTileA: 64 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 1 + enableLDSTrA: 0 + enableLDSTrB: 0 + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 33 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x224x128_MIHAxeqAl5GZyYpWyj9YBlHLiAot0B1CVx9Nf_5OHRC1Q= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 53248 + LdsInitCVgprs: false + LdsNumBytes: 53248 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 53248 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 224 + MacroTileA: 128 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 7 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 34 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x192x128_MIIlrnP4n1WnSi3fe0xoxepRfTuuLMMp-jgbnYdGnnvYE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 6] + MIWaveTileA: 4 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 192 + MacroTileA: 128 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 6 + ThreadTileA: 16 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 35 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x128_MI5bZ7yznefmkxpzzbB-XiEKlKRJu2AMeEgvXx60Tq5L0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 5 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 36 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x128_MI72iEPgzvyM2ZrWa9mJE7BjRqJtpIrgM81I_tGNEjK0c= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 37 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x128_MI1xf_ukgQ-Z-qNS4jmkabhCAr3mB7WAQFBVo0jxB_s6wM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 3 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 38 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x128_MI1j8uVA7i_pGx8ZCBrMR8ZhK3U6bp7J7YjtbYnnKdPxDI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 39 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x128_MI1M6uR2wC2h6sr2Hl9YrF8zhWGQ7RgoVsQ-TM7lCE6pPI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 22528 + LdsInitCVgprs: false + LdsNumBytes: 22528 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22528 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 1 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 40 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x256x128_MI1lw4esJJcKXjrN7F_gKTEASvr4FPioQKVPOZImphrvBY= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 49152 + LdsInitCVgprs: false + LdsNumBytes: 49152 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 49152 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 8] + MIWaveTileA: 3 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 256 + MacroTileA: 96 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 96 + NumGlobalWriteVectorsPerThread: 96 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 8 + ThreadTileA: 12 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 41 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x224x128_MI1JixbiZckG_HlXVPjjWZHNInPUwLf2xyuU7BTFa5yHI0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 7] + MIWaveTileA: 3 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 224 + MacroTileA: 96 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 84 + NumGlobalWriteVectorsPerThread: 84 + NumLoadsA: 3 + NumLoadsB: 7 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 7 + ThreadTileA: 12 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 42 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x128_MI1wIqbeUYowBvH47eY0D90voHkVaoBg2lPdbX2d68nIJ4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 3 + NumLoadsB: 6 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 43 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x128_MI1J-2M-3q4LSCntZjUh3v8nZH3qUcDC89rCJ0Fdq3iMeI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 80896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 80896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 3 + NumLoadsB: 5 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 44 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x128_MI1Vr1CG6o8iRf8BbhZJxSu0FVBXoX31LtlPuM8cs_YEXU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 4 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 45 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x128_MI16Cnahg_B3KmoGxZkTO1OYZLBTHOrwHNY1VQd2t5IrylI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 3 + NumLoadsB: 3 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 46 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x128_MI16W4yMmcRT_Rcp3ZNo2gSW3q4SJdd9i9DRRKcgqox59VE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 24576 + LdsInitCVgprs: false + LdsNumBytes: 24576 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 24576 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 3 + NumLoadsB: 2 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 47 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x128_MI16zNms4_iu6xgFQiq42WUbXrNnFNyslp04sbAKWAlTbog= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 20480 + LdsInitCVgprs: false + LdsNumBytes: 20480 + LdsNumElementsAlignedA: 15360 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 15360 + LdsOffsetB_Blk: 48128 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20480 + LdsOffsetMetadata_Blk: 48128 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 3 + NumLoadsB: 1 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 48 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x256x128_MI11aFcbk23grO4-hK8aBkA0Q2TJ_RzSnSefhCsshDFWE8= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 8] + MIWaveTileA: 2 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 256 + MacroTileA: 64 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 8 + ThreadTileA: 8 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 49 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x128_MI1BIQ9uHrC9E8FHJ9mzReo0log_q-mFe47Gvfw9lRTxKU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 7 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 50 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x128_MI1oWIDEj_bU2sIdW6WZSqKJI8Okx_mH0MwWvgKg4p0Wh0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 51 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x128_MI1DfS6ucItLGFYy3q0Zz-JZd7icG_ycHgazunawsT8Pns= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 5 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 52 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x128_MI1jK-MUAAqzxLFJYDxWtp_lRnOH6G_Sa3LS7gsDkzJ5ZI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 53 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x128_MI16qBSTIiXAL1b2k0UyETAi0pu4GSfWwKXX5o-ybzBtncE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 24576 + LdsInitCVgprs: false + LdsNumBytes: 24576 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 24576 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 3 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 54 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x128_MI16mOGETtdY_W1srxs4Mfgd04W-cV79aCR6raMD3HA6jWU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 18432 + LdsInitCVgprs: false + LdsNumBytes: 18432 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 18432 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 55 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x128_MI16wIclDOa00a_BcP3sOUDpDc_qFY9RvbDE777DV1a9O7k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 25600 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 9216 + LdsOffsetMetadata_Blk: 25600 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 1 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 56 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x128_MI1pg1aJvOuqnmTxe_-2e9_1S4uFGoAaPhgR-zs28ui024= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 38912 + LdsInitCVgprs: false + LdsNumBytes: 38912 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 38912 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 57 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x128_MI1bJJ5iziycwP81Vg_rFbnRobq-WE2Gw-XuD7kvGV58LU= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 40960 + LdsInitCVgprs: false + LdsNumBytes: 40960 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 35840 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 70656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 40960 + LdsOffsetMetadata_Blk: 70656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 1 + NumLoadsB: 7 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 58 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x128_MI1c60kFRwy-kwCbpVB3Be-Tp-X0igfgGU9wMtdCOAL5xM= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 32768 + LdsInitCVgprs: false + LdsNumBytes: 32768 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 32768 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 1 + NumLoadsB: 6 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 59 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x128_MI13_eugo-sWMKarMTBe1WI4hmzJdiGvToykcHBbj4WXow= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 25600 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 30720 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 1 + NumLoadsB: 5 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 60 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x128_MI1gSIhM4a60TQQmpGca3zPH_1C9YMOrBSb2Bu9JrxvD-k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 22528 + LdsInitCVgprs: false + LdsNumBytes: 22528 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 22528 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 4 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 61 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x128_MI16YH7g6N9l5w5imlP3Lp9mdDXwtO_ZtzNO9yUXMFVOjKQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 20480 + LdsInitCVgprs: false + LdsNumBytes: 20480 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 15360 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 37888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 20480 + LdsOffsetMetadata_Blk: 37888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 1 + NumLoadsB: 3 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 62 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x128_MI16IFZMDGDs3-q6koIU2xIB_XUAfzhmng9Mg_46GO3Rqf4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 30720 + LdsInitCVgprs: false + LdsNumBytes: 30720 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 1 + NumLoadsB: 2 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 63 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x128_MI16rLKaVqye5ZVlglTUO9ioHrZ6WB-GJ_Qs2RqP4S8ZIXE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 32 + LVCA: 8 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 5120 + LdsNumElementsAlignedB: 5120 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 16384 + LdsOffsetB: 5120 + LdsOffsetB_Blk: 21504 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 5120 + LdsOffsetMetadata_Blk: 21504 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 1 + NumLoadsB: 1 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 64 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_15_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x240x128_MIyoGz0G9DWmdNz6WjS2Bq6rCsCrR_mXM4td8eMkMYEDw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 73216 + LdsInitCVgprs: false + LdsNumBytes: 73216 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 38400 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 165888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73216 + LdsOffsetMetadata_Blk: 165888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 15] + MIWaveTileA: 4 + MIWaveTileB: 15 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 240 + MacroTileA: 256 + MacroTileB: 240 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 240 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 8 + NumLoadsB: 15 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 15 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 15 + ThreadTileA: 16 + ThreadTileB: 15 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 65 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_13_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x208x128_MIBnc7kWhy6AvjwbNc5adjziCfJiLAVAD1RzoNcS6OMZQ= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 68096 + LdsInitCVgprs: false + LdsNumBytes: 68096 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 33280 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 165888 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 68096 + LdsOffsetMetadata_Blk: 165888 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 13] + MIWaveTileA: 4 + MIWaveTileB: 13 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 208 + MacroTileA: 256 + MacroTileB: 208 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 52 + NumLoadsA: 8 + NumLoadsB: 13 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 13 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 13 + ThreadTileA: 16 + ThreadTileB: 13 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 66 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_11_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x176x128_MIF5A-1D1ZIPjJsJJ7ttt4AlfEFPQG0riur7Zv_qXG5YI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 128512 + LdsInitCVgprs: false + LdsNumBytes: 128512 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 28160 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 11] + MIWaveTileA: 4 + MIWaveTileB: 11 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 176 + MacroTileA: 256 + MacroTileB: 176 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 44 + NumLoadsA: 8 + NumLoadsB: 11 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 11 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 11 + ThreadTileA: 16 + ThreadTileB: 11 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 67 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_9_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x144x128_MIyRCs1CYgk6qoLPVzZz2698DWR4wMJ7whNVJ8JwJfnZc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 123392 + LdsInitCVgprs: false + LdsNumBytes: 123392 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 9] + MIWaveTileA: 4 + MIWaveTileB: 9 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 144 + MacroTileA: 256 + MacroTileB: 144 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 8 + NumLoadsB: 9 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 9 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 9 + ThreadTileA: 16 + ThreadTileB: 9 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 68 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x112x128_MIaVDafPaNDXaLtu56b1YkZsQUI1aLmxb3jgdu_dv65EA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 17920 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 7] + MIWaveTileA: 4 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 112 + MacroTileA: 256 + MacroTileB: 112 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 8 + NumLoadsB: 7 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 7 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 7 + ThreadTileA: 16 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 69 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x128_MI1jNVYmoC6fjSjbIADTf0f9l0Cs4TKifqjd6Xql0ehllk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 47616 + LdsInitCVgprs: false + LdsNumBytes: 47616 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 12800 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 47616 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 5 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 70 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x128_MI1eX_SljyDEcS5rCPMhk2caJwb3hhe7qqzmb_HRJ0i-U0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 7680 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 3 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 71 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB8_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB128_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x128_MI1fPZ5I9MI2WZKscvBb8L9C1UGdH6oze92UNxsd_rmjZ0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 8 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 32 + LSPB: 16 + LVCA: 8 + LVCB: 16 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 128 + LdsBytesNoAmax: 37376 + LdsInitCVgprs: false + LdsNumBytes: 37376 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 2560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37376 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 1 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 72 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT15_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT240x256x128_MI17whl_3k4Y1EhiqZzn7y-o-UTqlat-Sd3AsiBr6V3nc= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 73216 + LdsInitCVgprs: false + LdsNumBytes: 73216 + LdsNumElementsAlignedA: 38400 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 38400 + LdsOffsetB_Blk: 169472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73216 + LdsOffsetMetadata_Blk: 169472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [15, 4] + MIWaveTileA: 15 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 240 + MacroTile1: 256 + MacroTileA: 240 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 240 + NumGlobalWriteVectorsPerThread: 240 + NumLoadsA: 15 + NumLoadsB: 8 + NumLoadsPerpendicularA: 15 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 60 + ThreadTile1: 4 + ThreadTileA: 60 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 73 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT13_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT208x256x128_MIjK835T0h-3Ar1N2b4kmmt65KtYLTJhfaZH3LmR2auFA= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 68096 + LdsInitCVgprs: false + LdsNumBytes: 68096 + LdsNumElementsAlignedA: 33280 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33280 + LdsOffsetB_Blk: 164352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 68096 + LdsOffsetMetadata_Blk: 164352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [13, 4] + MIWaveTileA: 13 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 208 + MacroTile1: 256 + MacroTileA: 208 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 208 + NumGlobalWriteVectorsPerThread: 208 + NumLoadsA: 13 + NumLoadsB: 8 + NumLoadsPerpendicularA: 13 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 52 + ThreadTile1: 4 + ThreadTileA: 52 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 74 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT11_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT176x256x128_MIgsJynaoGp5Lr3OC9AUIiNlgAW9kkq7qPUmgK4ohOA_k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 128512 + LdsInitCVgprs: false + LdsNumBytes: 128512 + LdsNumElementsAlignedA: 28160 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 28160 + LdsOffsetB_Blk: 93696 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 28160 + LdsOffsetMetadata_Blk: 93696 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [11, 4] + MIWaveTileA: 11 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 176 + MacroTile1: 256 + MacroTileA: 176 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 176 + NumGlobalWriteVectorsPerThread: 176 + NumLoadsA: 11 + NumLoadsB: 8 + NumLoadsPerpendicularA: 11 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 44 + ThreadTile1: 4 + ThreadTileA: 44 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 75 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI16x16x1_SN_LDSB0_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT9_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT144x256x128_MI6ZUiH-qsm4LTRuwyb7uw6oTXUaPI3Vjpt7WHaUbBOfI= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 123392 + LdsInitCVgprs: false + LdsNumBytes: 123392 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 88576 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 23040 + LdsOffsetMetadata_Blk: 88576 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [9, 4] + MIWaveTileA: 9 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 144 + MacroTile1: 256 + MacroTileA: 144 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 144 + NumGlobalWriteVectorsPerThread: 144 + NumLoadsA: 9 + NumLoadsB: 8 + NumLoadsPerpendicularA: 9 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 36 + ThreadTile1: 4 + ThreadTileA: 36 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 76 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT112x256x128_MIiu6Dap7c-mOezayg2ePWEWJWIUNWwq_GUAWuc6asW8k= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 52736 + LdsInitCVgprs: false + LdsNumBytes: 52736 + LdsNumElementsAlignedA: 17920 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17920 + LdsOffsetB_Blk: 83456 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 52736 + LdsOffsetMetadata_Blk: 83456 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [7, 4] + MIWaveTileA: 7 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 112 + MacroTile1: 256 + MacroTileA: 112 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 112 + NumGlobalWriteVectorsPerThread: 112 + NumLoadsA: 7 + NumLoadsB: 8 + NumLoadsPerpendicularA: 7 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 4 + ThreadTileA: 28 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 77 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x128_MI1fyJI8fnAEv3sOPHcVPqOC2AU9kSqOftCPxastr6HZJE= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 47616 + LdsInitCVgprs: false + LdsNumBytes: 47616 + LdsNumElementsAlignedA: 12800 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 12800 + LdsOffsetB_Blk: 78336 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 47616 + LdsOffsetMetadata_Blk: 78336 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 8 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 78 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x128_MI1-Zo3r6x3T98ZMWTZ4H7LqLtk7xXmH_6zOjhPN09FdTk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 42496 + LdsInitCVgprs: false + LdsNumBytes: 42496 + LdsNumElementsAlignedA: 7680 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 7680 + LdsOffsetB_Blk: 73216 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 42496 + LdsOffsetMetadata_Blk: 73216 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 8 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 79 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA8_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA128_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS128_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x128_MI1wehOWUghk2A68IMMHO37hmuhc-kFrA6X9yfb9WRqhBg= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 128 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 8 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 128 + LSCB: 128 + LSPA: 16 + LSPB: 32 + LVCA: 16 + LVCB: 8 + LVPA: 2 + LVPB: 2 + LdsBlockSizePerPadA: 128 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 37376 + LdsInitCVgprs: false + LdsNumBytes: 37376 + LdsNumElementsAlignedA: 2560 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 2560 + LdsOffsetB_Blk: 68096 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 37376 + LdsOffsetMetadata_Blk: 68096 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 8 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 128 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 128 + _DepthUA: 128 + _DepthUB: 128 + _DepthUMetadata: 128 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 80 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA2048_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT8_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW8_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA8_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x32x256_MI1vCzuwEprjPexlyglCW3zRXsjX9xUmGeZzokhH60Z9yE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 8 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 2048 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 75776 + LdsInitCVgprs: false + LdsNumBytes: 75776 + LdsNumElementsAlignedA: 66560 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 66560 + LdsOffsetB_Blk: 197632 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 75776 + LdsOffsetMetadata_Blk: 197632 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [8, 1] + MIWaveTileA: 8 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 32 + MacroTileA: 256 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 2 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 8 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 32 + ThreadTile1: 1 + ThreadTileA: 32 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 8 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 81 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x64x256_MI1SiLGOArtDjYGdvuIvIxglRBR2PASF7znHipyq968J9k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 81920 + LdsInitCVgprs: false + LdsNumBytes: 81920 + LdsNumElementsAlignedA: 64512 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 64512 + LdsOffsetB_Blk: 195584 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81920 + LdsOffsetMetadata_Blk: 195584 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 2] + MIWaveTileA: 7 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 64 + MacroTileA: 224 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 56 + NumLoadsA: 14 + NumLoadsB: 4 + NumLoadsPerpendicularA: 14 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 2 + ThreadTileA: 28 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 82 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT7_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT224x32x256_MI1Atkrqghs5hZ7rFrRDMn1qRHk9P7RjRIX2jTHJaCZcIs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 64512 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 64512 + LdsOffsetB_Blk: 195584 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 195584 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [7, 1] + MIWaveTileA: 7 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 224 + MacroTile1: 32 + MacroTileA: 224 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 14 + NumLoadsB: 2 + NumLoadsPerpendicularA: 14 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 28 + ThreadTile1: 1 + ThreadTileA: 28 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 83 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x96x256_MI1H6kBMzYhDsiXY1vBEU2Ud8UhLYehCYuWeQaCulrUTHo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 3] + MIWaveTileA: 6 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 96 + MacroTileA: 192 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 12 + NumLoadsB: 6 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 3 + ThreadTileA: 24 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 84 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x64x256_MI153vNbJPjGrVniCPDOzS5UGzSK6NOLhTT3dYrLoo1YJ4= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 2] + MIWaveTileA: 6 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 64 + MacroTileA: 192 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 12 + NumLoadsB: 4 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 2 + ThreadTileA: 24 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 85 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT6_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT192x32x256_MI1y0BaulKQmTtKr2XSp5P3dnaOK0k5UYUxfkhT1HxC_rU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 117760 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 117760 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [6, 1] + MIWaveTileA: 6 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 192 + MacroTile1: 32 + MacroTileA: 192 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 2 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 24 + ThreadTile1: 1 + ThreadTileA: 24 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 86 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x128x256_MIUYoy0JGLYkrhIavz9gjog-Elk__Z0jYq-QXQEcN2PaA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 177152 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 177152 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 128 + MacroTileA: 160 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 10 + NumLoadsB: 8 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 87 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x96x256_MI1cUVPZBoZ-mtmE8sTWF_Az_FU-gB_6I8zhwfiN1pEaks= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 177152 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 177152 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 3] + MIWaveTileA: 5 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 96 + MacroTileA: 160 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 10 + NumLoadsB: 6 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 3 + ThreadTileA: 20 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 88 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x64x256_MI1EM3GcfaRjwnLLnXwjtnf8XZHB699buVut-TXk4ZjWyc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 63488 + LdsInitCVgprs: false + LdsNumBytes: 63488 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 63488 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 2] + MIWaveTileA: 5 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 64 + MacroTileA: 160 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 40 + NumLoadsA: 10 + NumLoadsB: 4 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 2 + ThreadTileA: 20 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 89 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT160x32x256_MI15N7cdBjXALKtpaYkT17ouFsMqBdqYm1rqUc2cC7GeLE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 46080 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 46080 + LdsOffsetB_Blk: 111616 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 111616 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [5, 1] + MIWaveTileA: 5 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 160 + MacroTile1: 32 + MacroTileA: 160 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 10 + NumLoadsB: 2 + NumLoadsPerpendicularA: 10 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 1 + ThreadTileA: 20 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 90 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x160x256_MIU14cbwZ_t9NC-mBYr7N5MljJmeoGwBb-WUmjyy7ILhU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 160 + MacroTileA: 128 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 8 + NumLoadsB: 10 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 91 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x128x256_MIM08mmLtqT14x3TxGefabTcO290EciaeH6GnjDeU8Bgo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 4] + MIWaveTileA: 4 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 128 + MacroTileA: 128 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 64 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 4 + ThreadTileA: 16 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 92 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x96x256_MI1OG3k0SIeaxxBYmDDXCQJ0v824MXDB2cvmAH14Io8mlM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 96 + MacroTileA: 128 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 8 + NumLoadsB: 6 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 93 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x64x256_MI1Ee_Jq2fBrVEHXaPp-qikzlaFXpjSm2wKDPvD-MKxLM0= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 2] + MIWaveTileA: 4 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 64 + MacroTileA: 128 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 2 + ThreadTileA: 16 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 94 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT128x32x256_MI13IeiqjCNSTMel5JBLjnHPiB9PE_JYWIIfXqePFminMc= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 128 + MacroTile1: 32 + MacroTileA: 128 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 95 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x192x256_MI1qUn-crPudL6dXZ6rCcKnkE9s4w-RcuNNqOhYdbIrR28= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 79872 + LdsInitCVgprs: false + LdsNumBytes: 79872 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 158720 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 79872 + LdsOffsetMetadata_Blk: 158720 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 6] + MIWaveTileA: 3 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 192 + MacroTileA: 96 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 72 + NumGlobalWriteVectorsPerThread: 72 + NumLoadsA: 6 + NumLoadsB: 12 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 6 + ThreadTileA: 12 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 96 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x160x256_MI18wp9zZc9oencMlWPaDY7BUc_iQgMxwF-RQOzHet7J3I= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 158720 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 158720 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 5] + MIWaveTileA: 3 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 160 + MacroTileA: 96 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 60 + NumGlobalWriteVectorsPerThread: 60 + NumLoadsA: 6 + NumLoadsB: 10 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 5 + ThreadTileA: 12 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 97 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x128x256_MI1LrtbILHycv46qRt6NxJYDFm4zT_v-o9V1TAKrHNeAeQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 128 + MacroTileA: 96 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 6 + NumLoadsB: 8 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 98 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x96x256_MI164Fx4NhdfXwLWoN6X-Afen5a5uy0a8LeKZgpjvani3CM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 3] + MIWaveTileA: 3 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 96 + MacroTileA: 96 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 36 + NumGlobalWriteVectorsPerThread: 36 + NumLoadsA: 6 + NumLoadsB: 6 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 3 + ThreadTileA: 12 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 99 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x64x256_MI16_Nc2lY4fwUC-cbMivJ1U_eNf_9vhMW0WYu6jNMKBzYI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 2] + MIWaveTileA: 3 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 64 + MacroTileA: 96 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 6 + NumLoadsB: 4 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 2 + ThreadTileA: 12 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 100 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x256_MI16X2E4Wx9NVLCyWp4uy3bGT7QFxR20eVc2o2HQJqwi8Hk= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 27648 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 27648 + LdsOffsetB_Blk: 93184 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 93184 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 6 + NumLoadsB: 2 + NumLoadsPerpendicularA: 6 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 101 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x224x256_MI1ehl3l0UKCFUW0CqlYWmD5Gv6DpebCzV_7KPnbR_5y4Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 81920 + LdsInitCVgprs: false + LdsNumBytes: 81920 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 64512 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81920 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 7] + MIWaveTileA: 2 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 224 + MacroTileA: 64 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 56 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 4 + NumLoadsB: 14 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 14 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 7 + ThreadTileA: 8 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 102 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x192x256_MI1PpwSxkO--tNKZnLh5u1Zqztz5GRs2qwDUtmWoQ32_0M= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 6] + MIWaveTileA: 2 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 192 + MacroTileA: 64 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 4 + NumLoadsB: 12 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 6 + ThreadTileA: 8 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 103 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x160x256_MI1BIcxM2OYKG0Jqn5GRpKyHvCNhj2aQceoHMvTm7lJo6Q= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 63488 + LdsInitCVgprs: false + LdsNumBytes: 63488 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 63488 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 5] + MIWaveTileA: 2 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 160 + MacroTileA: 64 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 40 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 4 + NumLoadsB: 10 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 5 + ThreadTileA: 8 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 104 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x128x256_MI1-eL7HedP-uoMH-4uLF-vatbGDwtmp3GZhw8IuLzDHvE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 4] + MIWaveTileA: 2 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 128 + MacroTileA: 64 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 4 + ThreadTileA: 8 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 105 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x96x256_MI16Tidk1Ou4UA72T_vnHnWAsymx18z-7WG5S7yV6EMzxAE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 45056 + LdsInitCVgprs: false + LdsNumBytes: 45056 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 45056 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 3] + MIWaveTileA: 2 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 96 + MacroTileA: 64 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 6 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 3 + ThreadTileA: 8 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 106 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x256_MI16P7nrnDgCqsPXJVFvUTHGSoy-i5STapWZDg4F8KHN2a8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 107 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x256_MI164FNCErQfeGGweRYwOlRLTlQQw-uihsMini7VG5MjqUs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 108 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB2048_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_8_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB8_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x256x256_MI1Weg33rA_lR82ctIL5Wv-agG74LvDUGtagD8ZNXCm6m8= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 2048 + LdsBytesNoAmax: 75776 + LdsInitCVgprs: false + LdsNumBytes: 75776 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 66560 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 140288 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 75776 + LdsOffsetMetadata_Blk: 140288 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 8] + MIWaveTileA: 1 + MIWaveTileB: 8 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 256 + MacroTileA: 32 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 32 + NumGlobalWriteVectorsPerThread: 32 + NumLoadsA: 2 + NumLoadsB: 16 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 8 + ThreadTileA: 4 + ThreadTileB: 8 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 8 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 109 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_7_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x224x256_MI1yjWOfHPT7EqX9rhhXejY8_eiJ07brJZxCVBlCXR9DBo= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 73728 + LdsInitCVgprs: false + LdsNumBytes: 73728 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 64512 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 140288 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 73728 + LdsOffsetMetadata_Blk: 140288 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 7] + MIWaveTileA: 1 + MIWaveTileB: 7 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 224 + MacroTileA: 32 + MacroTileB: 224 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 28 + NumGlobalWriteVectorsPerThread: 28 + NumLoadsA: 2 + NumLoadsB: 14 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 14 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 7 + ThreadTileA: 4 + ThreadTileB: 7 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 110 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_6_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x192x256_MI1pVP6pjfyPWPVeerMg8FY1EyiGjH4B_j7tc4lRUzRPvg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 61440 + LdsInitCVgprs: false + LdsNumBytes: 61440 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 61440 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 6] + MIWaveTileA: 1 + MIWaveTileB: 6 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 192 + MacroTileA: 32 + MacroTileB: 192 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 24 + NumGlobalWriteVectorsPerThread: 24 + NumLoadsA: 2 + NumLoadsB: 12 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 6 + ThreadTileA: 4 + ThreadTileB: 6 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 111 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x160x256_MI1q7mTED5UvVJrOPDvYFNPflDjUkX1okY5IJT3AZBddMM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 55296 + LdsInitCVgprs: false + LdsNumBytes: 55296 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 46080 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 55296 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 5] + MIWaveTileA: 1 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 160 + MacroTileA: 32 + MacroTileB: 160 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 20 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 2 + NumLoadsB: 10 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 10 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 5 + ThreadTileA: 4 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 112 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x128x256_MI1fFpYWYjPWsmrBYgoz9CIhz7OHVoysPvt9SgTel6GPWI= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 43008 + LdsInitCVgprs: false + LdsNumBytes: 43008 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43008 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 128 + MacroTileA: 32 + MacroTileB: 128 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 113 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x256_MI16mu07e3PYZ_jp2LQxX0lmvu4pmsw_265fxGpIWycVSjg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 36864 + LdsInitCVgprs: false + LdsNumBytes: 36864 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 27648 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 74752 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 36864 + LdsOffsetMetadata_Blk: 74752 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 2 + NumLoadsB: 6 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 6 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 114 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x256_MI162x9lNzvJGF2jhOPNNRp1Q3mFaUkUkcNWlM7ecLmXXfs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26624 + LdsInitCVgprs: false + LdsNumBytes: 26624 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26624 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 115 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x256_MI16-01AyvDsXfpTy-wvoK3K25-ZRKuFeCLIRsj5lVXzX_s= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 18432 + LdsInitCVgprs: false + LdsNumBytes: 18432 + LdsNumElementsAlignedA: 9216 + LdsNumElementsAlignedB: 9216 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 9216 + LdsOffsetB_Blk: 41984 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 18432 + LdsOffsetMetadata_Blk: 41984 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 116 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_5_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x80x256_MI1LDs52gy7NI4WuOMQ_idbNXBwBK86orsq1YVyOo4IvKs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 90624 + LdsInitCVgprs: false + LdsNumBytes: 90624 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 23040 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 90624 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 5] + MIWaveTileA: 4 + MIWaveTileB: 5 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 80 + MacroTileA: 256 + MacroTileB: 80 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 20 + NumLoadsA: 16 + NumLoadsB: 5 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 5 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 5 + ThreadTileA: 16 + ThreadTileB: 5 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 117 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x48x256_MI1NztuuZzeaShCMYTqz64dRhvSvIEetpxjBMHS12EujLs= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 81408 + LdsInitCVgprs: false + LdsNumBytes: 81408 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 13824 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81408 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 3] + MIWaveTileA: 4 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 48 + MacroTileA: 256 + MacroTileB: 48 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 16 + NumLoadsB: 3 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 3 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 3 + ThreadTileA: 16 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 118 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB256_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW4_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA4_VWB1_WSGRA0_WSGRB0_WS64_WG64_4_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT256x16x256_MI1-eS3GREuZbaF_WcdTrJ8tJN1VRBcjitHDgD1jEhWCcE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 4 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 256 + LdsBytesNoAmax: 72192 + LdsInitCVgprs: false + LdsNumBytes: 72192 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 4608 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 72192 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [4, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 256 + MacroTile1: 16 + MacroTileA: 256 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 1 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 1 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 4 + StreamK: 5 + SubGroup0: 16 + SubGroup1: 16 + SubGroupA: 16 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 4 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [64, 4, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 119 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT5_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT80x256x256_MI1ZLUtUYYiOHV_3Bx1CweJ56xkOj29nbS2TUKPiIsgWeM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 90624 + LdsInitCVgprs: false + LdsNumBytes: 90624 + LdsNumElementsAlignedA: 23040 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 23040 + LdsOffsetB_Blk: 154112 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 90624 + LdsOffsetMetadata_Blk: 154112 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [5, 4] + MIWaveTileA: 5 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 80 + MacroTile1: 256 + MacroTileA: 80 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 80 + NumGlobalWriteVectorsPerThread: 80 + NumLoadsA: 5 + NumLoadsB: 16 + NumLoadsPerpendicularA: 5 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 20 + ThreadTile1: 4 + ThreadTileA: 20 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 120 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT48x256x256_MI1QlOH2GDodFrn6A0rxCKHrFBaU_QWfv580nm9xWdZY0k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 81408 + LdsInitCVgprs: false + LdsNumBytes: 81408 + LdsNumElementsAlignedA: 13824 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 13824 + LdsOffsetB_Blk: 144896 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 81408 + LdsOffsetMetadata_Blk: 144896 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [3, 4] + MIWaveTileA: 3 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 48 + MacroTile1: 256 + MacroTileA: 48 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 48 + NumGlobalWriteVectorsPerThread: 48 + NumLoadsA: 3 + NumLoadsB: 16 + NumLoadsPerpendicularA: 3 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 4 + ThreadTileA: 12 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 121 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA256_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS256_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB4_WSGRA0_WSGRB0_WS64_WG16_16_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x256x256_MI1PZ1TyY_JO23lBPKX9VhACTOX37NNgYiBjf2A-tanW7w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 256 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 256 + LSCB: 256 + LSPA: 16 + LSPB: 16 + LVCA: 16 + LVCB: 16 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 256 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 72192 + LdsInitCVgprs: false + LdsNumBytes: 72192 + LdsNumElementsAlignedA: 4608 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 4608 + LdsOffsetB_Blk: 135680 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 72192 + LdsOffsetMetadata_Blk: 135680 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 4] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 256 + MacroTileA: 16 + MacroTileB: 256 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 16 + NumLoadsA: 1 + NumLoadsB: 16 + NumLoadsPerpendicularA: 1 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 64 + SubGroupA: 4 + SubGroupB: 64 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 4 + WavefrontSize: 64 + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 256 + _DepthUA: 256 + _DepthUB: 256 + _DepthUMetadata: 256 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 122 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT3_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT96x32x512_MI16Z6LYIKVcBKQJUXMfCTkH2UmyhUWm90vrVgFH5F5h_is= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 52224 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 52224 + LdsOffsetB_Blk: 183296 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 183296 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [3, 1] + MIWaveTileA: 3 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 96 + MacroTile1: 32 + MacroTileA: 96 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 12 + NumLoadsB: 4 + NumLoadsPerpendicularA: 12 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 12 + ThreadTile1: 1 + ThreadTileA: 12 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 123 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x64x512_MI16Gx0c38MGhJUAjADEf6sgUUCYEnSxT0HvEevn-TLzDBA= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 2] + MIWaveTileA: 2 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 64 + MacroTileA: 64 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 16 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 2 + ThreadTileA: 8 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 124 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW2_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA2_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x32x512_MI16-AdnHWdGaF-mVZCwCYUd2wq3b2fSNewc3-7rmRp2b3Y= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 2 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 32 + MacroTileA: 64 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 2 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 2 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 125 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_3_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x96x512_MI16tO84-u8CmDozhzOYFPjZx1aAnLl_RdoGaIlGdosGUzM= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 69632 + LdsInitCVgprs: false + LdsNumBytes: 69632 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 52224 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 148480 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 69632 + LdsOffsetMetadata_Blk: 148480 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 3] + MIWaveTileA: 1 + MIWaveTileB: 3 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 96 + MacroTileA: 32 + MacroTileB: 96 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 12 + NumGlobalWriteVectorsPerThread: 12 + NumLoadsA: 4 + NumLoadsB: 12 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 12 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 3 + ThreadTileA: 4 + ThreadTileB: 3 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 126 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB2_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x64x512_MI16BfEq8Qvc7y_UvNL_Jl4pNSs8RolDE0lSBE0kSa4q37k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 51200 + LdsInitCVgprs: false + LdsNumBytes: 51200 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 51200 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 64 + MacroTileA: 32 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 8 + NumGlobalWriteVectorsPerThread: 8 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 2 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 127 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x512_MI16nxXVLq-_LhX2N8-ggvlvGFb2CSQPdL0Jk31-3Yn8K8w= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 34816 + LdsInitCVgprs: false + LdsNumBytes: 34816 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 82944 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 34816 + LdsOffsetMetadata_Blk: 82944 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 4 + LoopUnroll: 512 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 128 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG32_8_1_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x32x1024_MI1Uh3YqlpEBhxKyjcdbTJSpHtEarJLrJfEIa0Mnxwsb5k= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 67584 + LdsInitCVgprs: false + LdsNumBytes: 67584 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 164864 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 67584 + LdsOffsetMetadata_Blk: 164864 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 1 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 8 + LoopUnroll: 1024 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [2, 2] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 32 + MacroTileA: 32 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 8 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 8 + SubGroup1: 32 + SubGroupA: 8 + SubGroupB: 32 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [32, 8, 1] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 129 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x512_MI16t0-mPljiEZtupRhWAFpnxBDrJen-FsvDD34Q7PN3s_4= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 17408 + LdsInitCVgprs: false + LdsNumBytes: 17408 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 17408 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 2 + NumLoadsB: 2 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 130 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x512_MI16CAJPr2P6olRKJqA2Fmiymyoh6JhX7nYYGS7ffsGJvmw= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 17408 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 17408 + LdsOffsetB_Blk: 50176 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 50176 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 2 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 131 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x512_MI16K_Na50sAdM91rhXevAlyiM9lrmobRF_PSYV6N76rQzk= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 34816 + LdsNumElementsAlignedB: 8704 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 34816 + LdsOffsetB_Blk: 100352 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 100352 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 8 + NumLoadsB: 2 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 2 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 132 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x512_MI16dVDH5F1pzH8NV-uymOAiVxprnNui5yf2Mp5t1_LqLTs= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 26112 + LdsInitCVgprs: false + LdsNumBytes: 26112 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 17408 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 32768 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 41472 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 26112 + LdsOffsetMetadata_Blk: 41472 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 2 + NumLoadsB: 4 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 133 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR0_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA512_LBSPPB512_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR0_PKA1_SIA3_SS1_SU0_SUM1_SUS512_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x512_MI16Uv-d7icfFxrZ-YUiRaTBsX3xXbc3HcmSMlEdtwh8Yb0= + CUCount: null + CUOccupancy: -1 + ClusterLocalRead: 0 + CodeObjectVersion: 4 + DepthU: 512 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 512 + LSCB: 512 + LSPA: 8 + LSPB: 8 + LVCA: 32 + LVCB: 32 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 512 + LdsBlockSizePerPadB: 512 + LdsBytesNoAmax: 43520 + LdsInitCVgprs: false + LdsNumBytes: 43520 + LdsNumElementsAlignedA: 8704 + LdsNumElementsAlignedB: 34816 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 8704 + LdsOffsetB_Blk: 74240 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 43520 + LdsOffsetMetadata_Blk: 74240 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 1 + LoopUnroll: 128 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 2 + NumLoadsB: 8 + NumLoadsPerpendicularA: 2 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + PrefetchLocalRead: 0 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 512 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 512 + _DepthUA: 512 + _DepthUB: 512 + _DepthUMetadata: 512 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 134 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x16x1024_MI1TNTLRLqQ0uhz1zq5XPt3rxttGE1MT5QoFsMWFmpwZGg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 33792 + LdsInitCVgprs: false + LdsNumBytes: 33792 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 33792 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 1] + MIWaveTileA: 1 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 16 + MacroTileA: 16 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 1 + NumGlobalWriteVectorsPerThread: 1 + NumLoadsA: 4 + NumLoadsB: 4 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 1 + ThreadTileA: 4 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 135 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT2_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT32x16x1024_MI13HHr0kFgQAqJ2TjtZ8tTohe6Udz-5u0CwkMq-_9renQ= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 33792 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 33792 + LdsOffsetB_Blk: 99328 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 99328 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [2, 1] + MIWaveTileA: 2 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 32 + MacroTile1: 16 + MacroTileA: 32 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 8 + NumLoadsB: 4 + NumLoadsPerpendicularA: 8 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 8 + ThreadTile1: 1 + ThreadTileA: 8 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 136 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT4_1_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT64x16x1024_MI1KAql7WR7THsIAKFHzmlcHQVpFofe0Vzt2xFE0wiTOfU= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84480 + LdsInitCVgprs: false + LdsNumBytes: 84480 + LdsNumElementsAlignedA: 67584 + LdsNumElementsAlignedB: 16896 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 67584 + LdsOffsetB_Blk: 198656 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84480 + LdsOffsetMetadata_Blk: 198656 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [4, 1] + MIWaveTileA: 4 + MIWaveTileB: 1 + MIWaveTileMetadata: 0 + MacroTile0: 64 + MacroTile1: 16 + MacroTileA: 64 + MacroTileB: 16 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 16 + NumLoadsB: 4 + NumLoadsPerpendicularA: 16 + NumLoadsPerpendicularB: 4 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 16 + ThreadTile1: 1 + ThreadTileA: 16 + ThreadTileB: 1 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 137 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_2_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x32x1024_MI1-LGVgPAQHxTiHAOxIZA3A-LLqgvwFhsRG4LoerQvYqg= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 50688 + LdsInitCVgprs: false + LdsNumBytes: 50688 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 33792 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 65536 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 82432 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 50688 + LdsOffsetMetadata_Blk: 82432 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 2] + MIWaveTileA: 1 + MIWaveTileB: 2 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 32 + MacroTileA: 16 + MacroTileB: 32 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 2 + NumGlobalWriteVectorsPerThread: 2 + NumLoadsA: 4 + NumLoadsB: 8 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 8 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 2 + ThreadTileA: 4 + ThreadTileB: 2 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +- SolutionIndex: 138 + KernelNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4 + SolutionNameMin: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI16x16x1_SN_LDSB1_AFC0_AFEM1_AFEM1_ASEM1_CLR1_CADS0_DTVA0_DTVB0_EPS0_FDSI0_GRPM1_GRVWA16_GRVWB16_GSU0_GSUAMB_GSUC0_GSUWGMRR0_GLS0_ISA950_IU1_K1_LBSPPA1024_LBSPPB1024_LBSPPM0_LPA32_LPB32_LPM0_LRVW16_LWPMn1_MIAV0_MIWT1_4_MO40_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SU0_SUM1_SUS1024_SPO0_SRVW0_SSO0_SVW1_SK5_SKXCCM0_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGRO0_VSn1_VWA1_VWB1_WSGRA0_WSGRB0_WS64_WG16_4_4_WGM0_WGMXCCn1_WGMXCCGn1 + 1LDSBuffer: 1 + ActivationFuncCall: false + AssignedDerivedParameters: true + AssignedProblemIndependentDerivedParameters: true + BaseName: Cijk_Alik_Bljk_F8B8HS_BH_BiasSH_HAS_SAB_SAV_UserArgs_MT16x64x1024_MI1uwD3q8H7ROxJTYYpP8cvtJiz3DeoGvbO5IiJaZpCFnE= + CUCount: null + CUOccupancy: -1 + CodeObjectVersion: 4 + DepthU: 1024 + DirectToLdsA: false + DirectToLdsB: false + EdgeType: ShiftPtr + EnableF32XEmulationLds: false + EnableF32XdlMathOp: false + EnableMatrixInstruction: true + ExpandPointerSwap: false + ExpertSchedulingMode: 0 + GlobalReadVectorWidthA: 16 + GlobalReadVectorWidthB: 16 + GlobalSplitU: 0 + GlobalWriteVectorWidth: 1 + GuaranteeNoPartialA: true + GuaranteeNoPartialB: true + GuaranteeNoPartialMetadata: true + ISA: [9, 5, 0] + InternalSupportParams: {KernArgsVersion: 2, SupportCustomStaggerU: true, SupportCustomWGM: true, + SupportUserGSU: false, UseUniversalArgs: true} + Kernel: true + LSCA: 1024 + LSCB: 1024 + LSPA: 4 + LSPB: 4 + LVCA: 64 + LVCB: 64 + LVPA: 1 + LVPB: 1 + LdsBlockSizePerPadA: 1024 + LdsBlockSizePerPadB: 1024 + LdsBytesNoAmax: 84480 + LdsInitCVgprs: false + LdsNumBytes: 84480 + LdsNumElementsAlignedA: 16896 + LdsNumElementsAlignedB: 67584 + LdsNumElementsAlignedMetadata: 0 + LdsOffsetA: 0 + LdsOffsetA_Blk: 131072 + LdsOffsetB: 16896 + LdsOffsetB_Blk: 147968 + LdsOffsetBias: 0 + LdsOffsetBiasGSU: 0 + LdsOffsetBiasNonGSU: 0 + LdsOffsetMetadata: 84480 + LdsOffsetMetadata_Blk: 147968 + LdsPadA: 32 + LdsPadB: 32 + LocalReadVectorWidth: 16 + LocalSplitU: 4 + LocalSplitUReuseLDS: 1 + LocalWriteUseSgprA: false + LocalWriteUseSgprB: false + LoopIters: 2 + LoopUnroll: 256 + MFMA_BF16_1K: false + MIBlock: [16, 16, 128, 1, 1, 1] + MIInputPerThread: 32 + MIInputPerThreadA: 32 + MIInputPerThreadB: 32 + MIInputPerThreadMetadata: 32 + MIOutputVectorWidth: 4 + MIRegPerOut: 1 + MIWaveGroup: [1, 1] + MIWaveTile: [1, 4] + MIWaveTileA: 1 + MIWaveTileB: 4 + MIWaveTileMetadata: 0 + MacroTile0: 16 + MacroTile1: 64 + MacroTileA: 16 + MacroTileB: 64 + MathClocksUnrolledLoop: 0 + MatrixInstB: 1 + MatrixInstBM: 1 + MatrixInstBN: 1 + MatrixInstK: 128 + MatrixInstM: 16 + MatrixInstN: 16 + MatrixInstruction: [16, 16, 128, 1] + MaxLDS: 163840 + MbskPrefetchMethod: 0 + NoLdsWriteCode: false + NoTailLoop: false + NonDTLTailLoopA: true + NonDTLTailLoopB: true + NumElementsPerThread: 4 + NumGlobalWriteVectorsPerThread: 4 + NumLoadsA: 4 + NumLoadsB: 16 + NumLoadsPerpendicularA: 4 + NumLoadsPerpendicularB: 16 + NumThreads: 256 + NumWaveSplitK: 1 + PackedC0IdxChars: [I] + PackedC0IndicesX: [0] + PackedC1IdxChars: [J] + PackedC1IndicesX: [1] + PrefetchGlobalRead: 2 + SourceSwap: true + StaggerU: 0 + StaggerUMapping: 1 + StaggerUStride: 1024 + StoreSwapAddr: false + StoreVectorWidth: 1 + StreamK: 5 + SubGroup0: 4 + SubGroup1: 16 + SubGroupA: 4 + SubGroupB: 16 + ThreadTile: [1, 1] + ThreadTile0: 4 + ThreadTile1: 4 + ThreadTileA: 4 + ThreadTileB: 4 + TransposeLDS: 1 + TransposeLDSMetadata: 1 + ULSGRODoubleG2L: 0 + UnrollMajorLDSA: true + UnrollMajorLDSB: true + UnrollMajorLDSMetadata: true + UseDotInstruction: false + UseF32XEmulation: false + UseSgprForGRO: 0 + Valid: true + VectorWidthA: 1 + VectorWidthB: 1 + WavefrontSize: 64 + WorkGroup: [16, 4, 4] + WorkGroupMapping: 0 + WorkGroupMappingXCC: -1 + WorkspaceCheck: [4, 0, 0] + _DepthU: 1024 + _DepthUA: 1024 + _DepthUB: 1024 + _DepthUMetadata: 1024 + _GlobalAccumulation: PartialsBuffer + _UseSgprForGRO: 0 + _VectorStore: 1 + _WorkspaceSizePerElemBias: 0 + _WorkspaceSizePerElemC: 4 + _staggerStrideShift: 0 + enableLDSTrA: false + enableLDSTrB: false + reorderGRInstForDTVA: false + reorderGRInstForDTVB: false + tailLoopOptA: false + tailLoopOptB: false +IndexOrder: null +ExactLogic: null +RangeLogic: null +TileSelectionIndices: null +PerfMetric: DeviceEfficiency +LibraryType: Prediction